unsloth/install.ps1
Leo Borcherding b6d5636cc0
fix/strix halo and windows AMD ROCm support (#5301)
* fix(studio): set HIP_VISIBLE_DEVICES in apply_gpu_ids for ROCm training workers

Training workers are spawned via multiprocessing spawn before detect_hardware()
runs, so IS_ROCM is still False. If the user never set HIP_VISIBLE_DEVICES in
their shell, _inherits_rocm_visibility is also False, leaving the worker with
only CUDA_VISIBLE_DEVICES set. On ROCm hosts the HIP runtime honors
HIP_VISIBLE_DEVICES over CUDA_VISIBLE_DEVICES, so the worker saw the full
device list and torch raised "no usable HIP accelerator" on some setups.

Fall back to probing torch.version.hip (a build-time attribute, safe to read
before GPU init) to detect ROCm when neither IS_ROCM nor inherited env vars
are available. Mirrors the existing fix in llama_cpp.py for llama-server
subprocess GPU pinning.

Fixes https://github.com/unslothai/unsloth/issues/5180

* test: tighten apply_gpu_ids ROCm fallback assertions

Replace loose OR chain with exact string matches, split into three
focused tests, and add a guard check for the try/except wrapper.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: detect ROCm unified memory (Strix Halo / AMD iGPU) via torch fallback

amd-smi on iGPUs with shared/unified memory (e.g. Radeon 8060S on Strix
Halo) reports only the dedicated VRAM slice (~512 MB) in its metric output,
so get_visible_gpu_utilization() was returning usable_gb ≈ 0.35 GB instead
of the full GTT pool (~128 GB).  torch.cuda.mem_get_info() already surfaces
the correct unified-pool size.

Add _reconcile_rocm_unified_memory(): after amd-smi returns a valid result
on a ROCm device, cross-check each device's vram_total_gb against
torch.cuda.mem_get_info().  When torch reports a larger total, replace the
amd-smi VRAM fields in-place.  No-op for discrete AMD GPUs where the two
sources agree.

Fixes: "Falling back to all visible GPUs -- model may not fit" on AMD iGPU
machines even when 100+ GB of unified memory is available.

* Apply unified-memory reconciliation in get_gpu_utilization too

The visible-GPU path was already corrected for AMD iGPUs with unified memory
(Strix Halo / Radeon 8060S), but get_gpu_utilization was still returning the
raw 512 MB amd-smi VRAM slice. Studio's /api/train/hardware endpoint and the
live GPU monitor read from this primary path, so users continued seeing the
wrong total even after auto_select_gpu_ids picked the right device.

Refactor to share the per-device correction:
  * _apply_unified_memory_correction(metrics, torch_info) -- the actual
    replacement logic, in-place on a single metrics dict.
  * _reconcile_rocm_unified_memory(...)                   -- multi-device,
    iterates utilization["devices"] (visible-GPU path).
  * _reconcile_primary_rocm_unified_memory(...)           -- single flat
    metrics dict (primary-GPU path), uses parent_visible_spec to pick the
    primary index, falls back to ordinal 0 when no visibility env is set.

get_gpu_utilization now calls the primary reconciler under IS_ROCM, so both
endpoints surface the real unified-memory pool on iGPUs while leaving
discrete AMD GPUs untouched (torch_total <= smi_total -> no replace).

* Use 'is not None' and log debug on torch.version.hip probe failures

Two small follow-ups to the apply_gpu_ids ROCm fallback:

1. Match detect_hardware()'s 'getattr(torch.version, "hip", None) is not None'
   form so the entire codebase has one canonical 'this torch was built with
   HIP' check. On every shipping torch wheel hip is either None or a non-empty
   version string, so the new form agrees with the old bool() form on every
   real install.

2. Log the probe failure at debug level instead of swallowing it silently.
   The broad 'except Exception' is intentional (we never want apply_gpu_ids
   to crash a worker over a probe), but the silent pass made it impossible
   to tell whether the fallback was firing or being skipped.

* fix(studio): honour HIP_VISIBLE_DEVICES in _get_parent_visible_gpu_spec before IS_ROCM is set

When a user has HIP_VISIBLE_DEVICES set in their shell (e.g. "1" to select
GPU 1) but detect_hardware() has not yet run in the Studio parent process,
IS_ROCM is still False.  _get_parent_visible_gpu_spec() was gated on IS_ROCM
so it fell through to CUDA_VISIBLE_DEVICES (unset), saw all physical GPUs,
and auto-selected index 0.  apply_gpu_ids then overwrote HIP_VISIBLE_DEVICES
with "0", making the intended GPU invisible to ROCm torch in the worker,
which triggered the "no usable HIP accelerator" error (issue #5180).

Apply the same _inherits_rocm_visibility pattern already used in
apply_gpu_ids: check for HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES in the
environment regardless of IS_ROCM so the correct GPU index is preserved.

* fix(install): harden AMD ROCm GPU detection for multi-GPU and env-filtered setups

The previous rocminfo awk pattern could miss discrete GPUs on machines
where HIP_VISIBLE_DEVICES/ROCR_VISIBLE_DEVICES is used to mask an
integrated GPU — the env vars filter rocminfo output but may not
propagate into the install script subprocess, causing detection to
fail entirely.

Two changes:
- Tighten rocminfo pattern from /gfx[0-9]/ && !/gfx000/ to
  /gfx[1-9][0-9]/ — simpler and correctly excludes the CPU agent
  (gfx000) without a negative lookahead
- Add sysfs KFD topology fallback: reads
  /sys/class/kfd/kfd/topology/nodes/*/gpu_id which is a kernel-level
  view unaffected by HIP_VISIBLE_DEVICES or ROCR_VISIBLE_DEVICES

Fixes detection failure reported in Discord by Chains (gfx1201 + iGPU
machine where env var exclusion of the iGPU caused rocminfo to return
no usable device).

* Fix KFD sysfs awk fallback to read properties file

The fallback added by this PR reads /sys/class/kfd/kfd/topology/nodes/*/gpu_id
files but matches the literal token 'gpu_id' against their content. Those
files contain only a single decimal value (e.g. '0' for CPU agents, '50432'
for GPU agents), so the regex never matches and 'found' stays 0, making the
fallback a no-op on every host. The properties file in the same directory
contains key/value lines like 'gpu_id 50432' which is what the existing awk
pattern expects.

Reproduced with a synthetic sysfs layout: against gpu_id files awk exits 1;
against properties files awk exits 0 when any node reports gpu_id > 0.

* fix(setup.ps1): detect AMD ROCm GPU on Windows, bring to parity with setup.sh

setup.ps1 only checked nvidia-smi and fell straight to "gpu: none" on AMD
machines. setup.sh already probed rocminfo/amd-smi/hipconfig/hipinfo.

Add three-tier detection mirroring install_llama_prebuilt.py's detect_host():
1. hipinfo: gcnArchName in output confirms a real HIP GPU (not just SDK)
2. amd-smi list: "GPU: <digit>" data rows as fallback
3. WMI Win32_VideoController: last resort -- detects AMD GPU even without
   HIP SDK, then guides user to install it rather than silently going CPU

Also corrects the "none" message to mention AMD ROCm alongside NVIDIA so
users with AMD hardware understand the requirement.

Fixes: rohit-style install where Strix Halo (Radeon 8060S) showed
"gpu: none" even with the HIP SDK present.

* fix(install.ps1): detect AMD ROCm GPU on Windows, bring to parity with setup.ps1

install.ps1 had the same nvidia-smi-only GPU detection as setup.ps1 before
the setup.ps1 fix. Applies the same three-tier AMD detection:
1. hipinfo: gcnArchName confirms real HIP GPU
2. amd-smi list: GPU data rows as fallback
3. WMI Win32_VideoController: detects AMD GPU without HIP SDK and guides
   user to install it

Fixes: install.ps1 showing "gpu: none" while setup.ps1 correctly showed
"AMD GPU detected" on the same machine (reported by rohit, RX 7600 XT).

* fix(install.ps1): suppress 'No NVIDIA GPU detected' when AMD GPU is present

* feat: add Windows AMD ROCm PyTorch wheel installation

install_python_stack.py:
- Add _ROCM_WINDOWS_WHEEL_BASE and _ROCM_WINDOWS_RELEASES constants
  pointing to AMD repo.radeon.com (ROCm 7.2 -> torch 2.9.1+rocm7.2.1)
- Extend _ensure_rocm_torch() with a Windows branch: detects ROCm via
  _has_rocm_gpu() / _detect_rocm_version(), requires Python 3.12 (cp312
  is the only ABI AMD publishes for Windows), installs the direct wheel
  URL from repo.radeon.com

install.ps1:
- Capture ROCmVersion during AMD detection via hipconfig --version /
  amd-smi version (needed for wheel URL selection)
- After Get-TorchIndexUrl, add an AMD wheel override block: when HasROCm
  and Python 3.12 detected, set ROCmTorchWheelUrl to AMD wheel URL
- Expand torch install branch to handle ROCmTorchWheelUrl with
  uv pip install --force-reinstall --no-cache-dir

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: also install torchvision and torchaudio from AMD Windows repo

AMD publishes matching torchvision-0.24.1+rocm7.2.1 and
torchaudio-2.9.1+rocm7.2.1 cp312 wheels at the same repo.radeon.com
release folder. Install all three in both install.ps1 and
install_python_stack.py Windows ROCm path.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* feat: add ROCm 7.1.1 Windows wheel mapping

AMD uses a different version string for 7.1.1 wheels:
2.9.0+rocmsdk20251116 (date-tagged) instead of +rocm7.1.1.
Adds the 7.1.1 release folder to both install.ps1 and
install_python_stack.py so users with ROCm 7.1 get ROCm
torch instead of falling back to CPU.

* fix: install rocm_sdk_core and rocm_sdk_libraries_custom alongside torch

The AMD Windows torch wheels declare rocm[libraries]==<ver> as a hard
dependency. Without installing rocm_sdk_core and rocm_sdk_libraries_custom
from the same AMD release folder, uv cannot resolve the dependency and
fails with 'No solution found'. Include all 5 wheels in one install call.

* fix: expand ROCm wheel array to scalars for Invoke-InstallCommand

@array splatting inside a scriptblock only works when the native command
is prefixed with '&'. Invoke-InstallCommand uses '& $Command' to run the
block, so @ROCmAllWheelUrls was not being expanded. Extract to scalar
variables $rw0-$rw4 which are captured correctly by the closure.

* fix: use --no-deps for AMD Windows torch wheel install

uv's resolver looks up rocm[libraries]==0.1.dev0 on PyPI during
dependency resolution before downloading any wheels, and fails because
the package doesn't exist on PyPI. --no-deps skips resolution entirely
and installs all 5 AMD wheels directly. The GPU runtime dependency is
satisfied by the HIP SDK, not a Python package.

* fix: setup.ps1 and install_python_stack.py now install ROCm torch on Windows

setup.ps1 was always setting CuTag='cpu' for non-NVIDIA hosts and installing
cpu-only PyTorch, overwriting the ROCm torch installed by install.ps1.
Adds the same AMD wheel selection logic (ROCm version detection, Python 3.12
check, 5-wheel install with --no-deps) to setup.ps1's torch install block.

install_python_stack.py: remove IS_WINDOWS guard from _ensure_rocm_torch()
call site so the Windows path in _ensure_rocm_torch() is reachable during
'unsloth studio update' as well.

* fix: suppress manual-install warning when ROCm torch already present; fix progress counter

- Gate the 'must be installed manually' warning on torch.version.hip being empty
  so it doesn't fire when our ROCm torch install succeeded
- Update _TOTAL counter to include the 3 ROCm steps on Windows now that
  _ensure_rocm_torch() is called there (fixes 10/9 display)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* feat: add rocm step display in setup.ps1; fix warning and progress counter

- Add 'rocm' step after 'cuda' in setup.ps1 showing ROCm version or HIP SDK missing
- Move ROCm version detection up to GPU detection block so it's available early
- Suppress 'must be installed manually' warning when torch.version.hip is set
- Fix _TOTAL counter to include ROCm steps on Windows (fixes 10/9 display)

* fix: detect AMD SDK ROCm torch via __version__ when torch.version.hip is unset

AMD's repo.radeon.com wheels (e.g. 2.9.0+rocmsdk20251116) do not set
torch.version.hip, leaving it None. All three probes that relied solely on
torch.version.hip now also check for 'rocm' in torch.__version__.lower():

- hardware.py detect_hardware(): IS_ROCM was never set, causing the studio
  to report 'Hardware detected: CPU' even after AMD wheels were installed
  and HIP DLLs were on PATH.
- install_python_stack.py _ensure_rocm_torch(): skip-if-already-installed
  probe would always reinstall on subsequent runs.
- install_python_stack.py Windows AMD warning: suppression check always
  failed, so the 'must be installed manually' note kept appearing after
  a successful AMD wheel install.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* perf: drop --no-cache-dir from AMD ROCm torch wheel installs

uv caches downloaded wheels by default; passing --no-cache-dir forced a
full redownload of the ~2 GB torch wheel on every install run. CUDA installs
never had this flag -- AMD was the only path affected.

* fix: use install-state flag instead of subprocess probe for AMD Windows warning

Replace the subprocess torch probe in the post-install warning block with a
module-level _rocm_windows_torch_installed flag set by _ensure_rocm_torch().
Subprocess re-import of torch is unnecessary and fragile -- the install
function already knows whether it succeeded.

* fix: hoist global declaration to top of _ensure_rocm_torch

Python requires the global statement to appear before any assignment
to the variable within a function. Moving it to the function top fixes
the SyntaxError on line 354.

* fix: pass AMD torch install status via env var to suppress false warning

setup.ps1 now sets UNSLOTH_ROCM_TORCH_INSTALLED=1 after a successful AMD
wheel install. install_python_stack.py reads this at the top of
_ensure_rocm_torch() to skip both the subprocess probe and the warning --
no re-import of torch needed, and the warning message now correctly says
'could not be auto-installed' rather than 'must be installed manually'.

* fix: register ROCm DLL directory before torch import on Windows

Python 3.8+ ignores PATH for extension DLL loading on Windows; amdhip64.dll
and other HIP runtime DLLs must be registered via os.add_dll_directory().
Without this, torch.cuda.is_available() always returns False on AMD ROCm
Windows even when HIP_PATH is correctly set in system environment variables.

Reads HIP_PATH / ROCM_PATH env vars first, then falls back to scanning
common ROCm install roots (C:\Program Files\AMD\ROCm, F:\ROCm, C:\ROCm).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: remove hardcoded non-standard ROCm paths from DLL directory scan

Only use HIP_PATH/ROCM_PATH (set by AMD installer) and the standard
C:\Program Files\AMD\ROCm\<version>\bin location. Custom drive paths
like F:\ROCm are user-specific and should not be hardcoded.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: prevent torchao overrides step from overwriting AMD ROCm torch

torchao==0.14.0 in overrides.txt declares torch as a dependency. Without
--no-deps, uv resolves torch from PyPI and installs 2.11.0+cpu on top of
the AMD ROCm wheels (2.9.0+rocmsdk20251116). This was the root cause of
'Hardware detected: CPU' -- the AMD wheels were installed but then
immediately overwritten by the overrides step.

When _rocm_windows_torch_installed is True, add --no-deps to the overrides
pip_install call so torchao is installed without pulling in CPU torch.

* fix: add rocm_sdk namespace tarball to Windows ROCm wheel installs

torch/_rocm_init.py calls `import rocm_sdk` at startup, which requires
the rocm namespace tarball (rocm-*.tar.gz) in addition to the SDK wheel
packages. This tarball was missing from both install.ps1 and setup.ps1,
causing ModuleNotFoundError on first torch import.

- Add rocm-0.1.dev0.tar.gz to ROCm 7.1.1 install (provides rocm_sdk namespace)
- Add rocm-7.2.1.tar.gz + rocm_sdk_devel to ROCm 7.2.1 install
- Install tarball in a dedicated step before main SDK/torch wheels
- Switch to @array splatting in install.ps1 scriptblock for dynamic wheel count
- Remove --no-cache-dir from Python-side ROCm wheel install (prevents ~2GB redownload)

* feat: enable ROCm 7.2 torch install + warn on gfx1151 with ROCm < 7.2

Chigoma333 (AMD Radeon 8060S / gfx1151, Strix Halo) confirmed that ROCm
7.1 segfaults when tensors are moved to GPU, but ROCm 7.2 + torch
2.11.0+rocm7.2 works fully including training.

Changes:
- Uncomment (7,2): "rocm7.2" in _ROCM_TORCH_INDEX (was blocked by <2.11.0)
- Add _ROCM_TORCH_PKG_SPECS dict with per-tag version bounds:
  rocm7.2 → torch>=2.11.0,<2.12.0; all older tags → <2.11.0
- Add _detect_amd_gfx_codes() helper that parses rocminfo output
- Warn on gfx1151/gfx1150 (Strix Halo) when ROCm < 7.2 is installed,
  pointing users at the known segfault and recommending upgrade
- install.sh get_torch_index_url(): enable rocm7.2 case (previously capped
  to rocm7.1), cap unknown future tags to rocm7.2
- install.sh: override TORCH_CONSTRAINT to >=2.11.0,<2.12.0 when rocm7.2
  index is selected, so pip can actually resolve torch 2.11.0

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: prefer Python 3.12 for AMD ROCm users when 3.13 is also installed

After GPU detection, if ROCm HIP SDK is found and the selected Python
is not 3.12, run a second pass to locate a 3.12 install via py.exe and
PATH (catches uv-managed installs). Switch $DetectedPython to 3.12 so
the venv is created with a compatible interpreter for the cp312-only AMD
Windows torch wheels.

NVIDIA and Intel GPU paths are unaffected -- the re-detection block only
runs when $HasROCm is true.

Fixes: #5301

* fix: also check uv-managed Python 3.12 for AMD ROCm #5301

* fix: hide amd-smi console popups on Windows, guard torch.distributed.is_initialized for ROCm #5301

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: suppress remaining console popups on Windows, patch torch.distributed.is_initialized for ROCm #5301

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: stub all missing torch.distributed attrs for ROCm Windows wheel #5301

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: inject torch.distributed stub when C backend missing in ROCm Windows wheel #5301

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(rocm/windows): pre-stub torch._C._distributed_c10d + raise amd-smi timeout

Two fixes for Windows ROCm regressions reported by electroglyph on #5301:

1. worker.py — torch.distributed stub now fires unconditionally on Windows
   The previous stub only injected sys.modules in the except branch, meaning
   it was silently skipped when `import torch.distributed` happened to succeed
   (the C backend is lazily resolved).  The crash then hit later when
   transformers/trl triggered the lazy load.  Fix: on win32 we pre-populate
   sys.modules['torch._C._distributed_c10d'] AND set the attribute on the
   torch._C extension module *before* attempting the import, covering both
   the early-ImportError and lazy-load failure modes.

2. amd.py — increase amd-smi timeout from 5 s to 30 s on Windows (10 s Linux)
   amd-smi on Windows must cold-init the ROCm runtime on first invocation;
   5 s was consistently too short, producing repeated 'Command timed out'
   warnings in the server log.  30 s gives enough headroom without blocking
   indefinitely on broken installs.

3. install.ps1 — widen Python 3.12 enforcement to ROCmGpuLabel (WMI-only path)
   Users whose HIP SDK is not on PATH were detected via WMI but not switched
   to Python 3.12 before the install started, causing a second pass.  Guard
   now fires on (HasROCm -or ROCmGpuLabel).

* fix(rocm): guard c10d stub, fix TorchIndexFamily for 7.1, clean dead code + comments

- worker.py: wrap c10d stub injection in `if _c10d_key not in sys.modules` so
  Windows NVIDIA users with a real torch.distributed are never affected
- install.ps1: fix Get-TauriTorchIndexFamily receiving hardcoded "rocm7.2"
  even when ROCm 7.1 wheels are installed; now branches on $ROCmVersion
- main.py: remove dead `import ctypes as _ctypes` (ctypes is never called)
- hardware.py, install_python_stack.py, worker.py, install.ps1: shorten
  verbose multi-line comment blocks throughout
- tests: update 4 stale assertions that expected rocm7.2 to be absent/capped

* fix(tests): match windows AMD warning assertion to actual source string

* chore: trim verbose comment blocks across all ROCm-related files

* fix: guard reconcile call against None numeric_ids; add torchvision lower bounds

* fix(install.ps1): recreate venv with Python 3.12 after ROCm switch

Venv was created with 3.13 before GPU detection ran; switching
$DetectedPython to 3.12 had no effect since $VenvPython still
pointed to the 3.13 interpreter inside the already-created venv.

* ux: detect AMD GPU before Python selection to avoid double venv creation

- Early hipinfo + WMI probe runs before Find-CompatiblePython so Python
  3.12 is selected upfront when AMD is detected; venv is now created
  exactly once instead of 3.13 then immediately 3.12.
- Post-venv recreation block replaced with a simple warning for the rare
  case where AMD was missed by the early probe.
- setup.ps1: show venv's actual Python version (e.g. 3.12) instead of
  the system Python found by the pre-activation search (was showing 3.13).

* fix(rocm/win): auto-stub all _distributed_c10d symbols via PEP-562 __getattr__

The bare ModuleType stub caused ImportError when torch._dynamo was imported
(triggered by trainer.py accessing torch._dynamo.config at load time).
torch._dynamo pulls in torch.distributed.fsdp._flat_param which does:
  from torch._C._distributed_c10d import FakeProcessGroup
and potentially other symbols. Adding module __getattr__ auto-creates a
stub class for any missing symbol so all such imports succeed without
enumerating every individual symbol. Applied to both the primary stub
and the fallback stub in the except branch.

* chore: trim c10d stub comment

* fix(rocm/win): auto-stub missing torch.distributed attrs (Store, ProcessGroup, …)

* fix(rocm/win): pre-stub fsdp submodules in sys.modules; fix __getattr__ subpackage clash

* feat(rocm/win): arch-aware wheel selector always picks newest ROCm release

Replace HIP-SDK-version-gated wheel selection with GPU arch-based logic.
Select-ROCmWheelRelease (PS) and _select_windows_rocm_release (Python) map
gcnArchName → minimum ROCm version, then pick the newest available release
that satisfies it (currently always rocm-rel-7.2.1 for any supported GPU).
Wheels bundle their own ROCm runtime so the installed HIP SDK 7.1 does not
prevent using 7.2.1 wheels on gfx1200 (RX 9060 XT) and similar RDNA 4 GPUs.

Also installs the bitsandbytes Windows ROCm continuous-release wheel and sets
BNB_ROCM_VERSION=72 in worker.py before ML imports so bnb loads the
libbitsandbytes_rocm72.dll that ships in that wheel.

* fix(rocm/win): stub class metaclass for ProcessGroup.BackendType; amd-smi circuit breaker

torchao.float8.inference accesses ProcessGroup.BackendType as a class-level
attribute.  Plain type() stubs have no __getattr__ on the metaclass so this
raises AttributeError.  Introduce _StubClassMeta whose __getattr__ returns
child stub classes, fixing the torchao import chain.

Add an amd-smi circuit breaker in amd.py: after 3 consecutive failures the
module stops spawning the process, eliminating the repeated Windows UAC /
DiskPart elevation prompts caused by polling a non-functional amd-smi.

Also guard BNB_ROCM_VERSION=72 behind a DLL existence check so bitsandbytes
fails with its own detection message rather than a harder "DLL not found" when
the Windows ROCm bnb wheel is not yet installed.

* fix: stub __members__ so torchao float8 enum check doesn't crash on ROCm Windows

torchao.float8.inference accesses ProcessGroup.BackendType.__members__
expecting a Python Enum registry dict. _StubClassMeta.__getattr__ was
blocking all dunder attributes, causing AttributeError. Return {} for
__members__ specifically so the isinstance/iteration checks pass cleanly.

* fix: stub distributed tensor/functional_collectives to prevent missing C++ op crash on ROCm Windows

torch._dynamo.trace_rules eagerly loads torch.distributed.tensor at import
time, which pulls in _functional_collectives.py. That file registers Meta
kernels for _c10d_functional C++ ops, but those ops are only registered
by torch._C._distributed_c10d — a C extension absent from ROCm Windows
wheels. Pre-stubbing the affected modules in sys.modules prevents the real
import chain from running and avoids the "operator does not exist" crash.

* fix: give mod stubs __path__ and pre-stub _tensor to fix 'not a package' import error

_make_mod_stub now sets __path__=[] so Python treats stub modules as
packages. Without it, any import of a submodule raises "is not a package".
Also pre-stub torch.distributed._tensor and its submodules so that
_tensor/__init__.py (which re-exports from torch.distributed.tensor) never
runs and torchao's `from torch.distributed._tensor import DTensor` gets a
harmless stub instead of crashing.

* fix: stub torch.ops._c10d_functional namespace with hashable op sentinels

torchao.dtypes.nf4tensor uses _c10d_functional ops as dict keys at import
time (all_gather_into_tensor.default, wait_tensor.default) and
torch.ops.c10d.scatter_.default. None of these ops are registered on ROCm
Windows because torch._C._distributed_c10d (the C extension) doesn't ship.
Replace the whole _c10d_functional namespace with a custom stub whose ops
return hashable .default objects, so dict-key construction doesn't crash.
Also inject a scatter_ stub into torch.ops.c10d if it's missing.

* fix: stub entire torchao package on ROCm Windows instead of individual ops

torchao is not supported on ROCm Windows and its import chain transitively
requires torch._C._distributed_c10d (absent from the ROCm Windows wheel).
Rather than stub each missing op one by one, stub the whole torchao package
upfront. Unsloth uses bitsandbytes for quantization, not torchao, so this
has no functional impact. transformers gracefully handles an importable-but-
empty torchao by disabling TorchAoHfQuantizer.

* fix: set __spec__ on mod stubs so importlib.util.find_spec doesn't raise

Manually-injected sys.modules entries have __spec__=None by default.
importlib.util.find_spec() raises ValueError when it finds a module in
sys.modules with __spec__=None (transformers.utils.import_utils hits this
when checking if torchao is available). Give every stub a minimal
ModuleSpec(name, loader=None, is_package=True) to satisfy find_spec.

* fix: add meta path finder to auto-stub subpackages of stub modules

`import torchao.prototype` goes through the import machinery, not
__getattr__, so an empty __path__ means ModuleNotFoundError. Rather than
list every submodule explicitly, register a MetaPathFinder that intercepts
any import whose parent is one of our stubs (detected by loader=None in the
parent's ModuleSpec). Real installed packages always have a SourceFileLoader
so they are never intercepted. Also register child stubs in sys.modules
from __getattr__ as a belt-and-suspenders measure.

* fix: use _unsloth_stub sentinel instead of loader=None for stub detection

The import machinery overwrites module.__spec__ with the spec returned by
find_spec (which has loader=_StubSubpackageLoader, not None), so the
loader=None check broke for second-level subpackages. Switch to a custom
_unsloth_stub object identity sentinel set directly on each stub module --
it survives __spec__ being replaced and correctly identifies stubs at any
depth (torchao.prototype.safetensors, etc.).

* refactor(rocm/win): switch to repo.amd.com arch-aware index, remove stubs

AMD recommends repo.amd.com/rocm/whl/{arch}/ as the Windows ROCm wheel
source. These wheels bundle their own ROCm runtime, support all Python
versions (not just cp312), and include the full torch._C extension set
(including _distributed_c10d) that the old repo.radeon.com wheel omitted.

Changes:
- install.ps1: remove Select-ROCmWheelRelease + hardcoded cp312 wheel
  URLs; remove Python 3.12 forced-preference logic; install via
  --index-url repo.amd.com/rocm/whl/{arch-family}/
- studio/setup.ps1: same -- remove Select-ROCmWheelRelease, switch to
  repo.amd.com arch-aware index URL
- studio/install_python_stack.py: replace _ROCM_WINDOWS_RELEASES /
  _select_windows_rocm_release with _windows_rocm_index_url() using the
  _GFX_TO_AMD_INDEX_ARCH map; drop Python 3.12 restriction
- studio/backend/core/training/worker.py: remove all stub machinery
  (_make_mod_stub, _StubSubpackageFinder, _StubSubpackageLoader,
  _StubClassMeta, torchao/fsdp/dtensor stubs, _c10d_functional ops
  stubs, BNB DLL detection) -- no longer needed with new wheel source

* fix(rocm/win): restore _distributed_c10d + torchao stubs; fix BNB install

repo.amd.com torch wheels also omit torch._C._distributed_c10d on Windows
(RCCL is not shipped on Windows). torch/distributed/__init__.py imports
from it unconditionally at module level, so the stub must land in
sys.modules before any torch.distributed import.

torchao (pulled in by transformers.quantizers) walks
torchao.float8.distributed_utils -> torch.distributed._functional_collectives
-> distributed_c10d at import time. Stubbing torchao up-front short-circuits
that chain.

worker.py:
- Restore _make_mod_stub / _StubSubpackageFinder / _StubSubpackageLoader
- Restore _StubClassMeta for ProcessGroup.BackendType attribute access
- Restore _distributed_c10d stub with __getattr__ (Windows only)
- Restore torchao stubs (5 modules, Windows only)

install_python_stack.py:
- BNB AMD wheel install was inside the early-return branch that fires when
  torch is already a ROCm build (installed by install.ps1). Move BNB install
  outside that branch so it always runs on Windows ROCm — the PyPI
  bitsandbytes has only CUDA DLLs and fails to load on ROCm.

* worker: remove _distributed_c10d stub; stub only torchao

The installed torch/distributed/__init__.py from repo.amd.com
(torch==2.10.0+rocm7.12.0) is now properly guarded with
`if is_available():`, so `import torch.distributed` alone is safe.

The crash only comes via torchao's import chain:
  torchao.float8.distributed_utils
    → torch.distributed._functional_collectives (unguarded import)
    → torch.distributed.distributed_c10d
    → torch._C._distributed_c10d  ← absent on Windows ROCm

Stubbing torchao short-circuits the chain entirely. No need to stub
_distributed_c10d. Remove _StubClassMeta and the _c10d stub block;
keep only _make_mod_stub + _StubSubpackageFinder + torchao seeds.

* fix: BNB AMD wheel skipped + torch.compile segfault on Windows ROCm

install_python_stack.py: the UNSLOTH_ROCM_TORCH_INSTALLED=1 early-return
path (set by setup.ps1 when it installed torch itself) returned before
ever reaching the AMD BNB prerelease wheel install.  The PyPI
bitsandbytes==0.49.x ships only CUDA DLLs, so loading it on ROCm fails
with "libbitsandbytes_rocm72.dll not found".  Now installs the AMD
Windows BNB wheel before returning on that path too.

worker.py: torch._grouped_mm crashes on gfx1200 (null HIP kernel pointer,
0xC0000005) when torch.compile's JitDecomp system dispatches it during
the first forward pass.  Detect Windows ROCm via torch.version.hip
(already in sys.modules from section 1e) and set TORCHDYNAMO_DISABLE=1
to bypass the broken kernel dispatch.

* fix: BNB AMD wheel install fails uv wheel filename check

The bitsandbytes continuous-release wheel is intentionally mismatched:
filename encodes 1.33.7.preview (= 1.33.7rc0 in PEP 440) but wheel
metadata reports 0.50.0.dev0.  uv rejects this by default.

Introduce _install_bnb_windows_rocm() helper that sets
UV_SKIP_WHEEL_FILENAME_CHECK=1 only for this specific install, then
restores the previous env value.  Both BNB install call sites (the
UNSLOTH_ROCM_TORCH_INSTALLED early-return path and the normal Windows
ROCm path) now use this helper.

* worker: patch _grouped_mm CUDA dispatch on Windows ROCm (gfx1200 null kernel)

TORCHDYNAMO_DISABLE=1 stopped the compiler frontend but not the autograd
JitDecomp system, which also dispatches _grouped_mm and hits the same
null HIP kernel crash (0xC0000005).

Verified that torch.library.Library("aten","IMPL").impl("_grouped_mm", fn,
"CUDA") successfully overrides the broken HIP kernel with a Python mm
fallback on torch==2.10.0+rocm7.12.0.

Schema: _grouped_mm(Tensor self, Tensor mat2, Tensor? offs=None,
                    Tensor? bias=None, ScalarType? out_dtype=None) -> Tensor

The fallback handles both the simple case (offs=None → torch.mm) and the
grouped case (offs provided → split self by offsets, multiply each group
against the corresponding slice of mat2, then cat results).

Keep _WINDOWS_ROCM_GROUPED_MM_LIB alive at function scope to prevent the
C++ dispatch registration from being freed by GC.

* worker: fix torchao stub — return stub classes not modules for isinstance()

peft/tuners/lora/torchao.py does:
  from torchao.dtypes import AffineQuantizedTensor, LinearActivationQuantizedTensor
  isinstance(weight, (AffineQuantizedTensor, LinearActivationQuantizedTensor))

The stub __getattr__ was returning stub modules, which isinstance() rejects
with "arg 2 must be a type, a tuple of types, or a union".

Add _StubTypeMeta metaclass whose __instancecheck__ always returns False,
and _make_stub_type() to create stub classes via it. Change _make_mod_stub
__getattr__ to return stub classes instead of stub modules for leaf
attribute access, so isinstance() gets a valid type and returns False.

_StubSubpackageFinder still handles import-style subpackage creation
(those still need module objects in sys.modules); __getattr__ only fires
for from-import or direct attribute access, which are the isinstance paths.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* tests: add coverage for Windows ROCm install paths and worker patches

Add conftest.py to fix pre-existing sys.path issue that prevented
test_rocm_support.py from running at all (install_python_stack.py
imports from backend.utils.wheel_utils which needs studio/ on sys.path).

New test classes cover everything added in this session:
- TestWindowsRocmIndexUrl: arch → AMD pip index URL mapping (gfx120X-all,
  gfx1151, gfx1150, gfx110X-all, unknown → None, trailing slash)
- TestDetectWindowsGfxArch: hipinfo output parsing, missing/timeout/bad
  returncode/no-gcnArchName paths
- TestInstallBnbWindowsRocm: UV_SKIP_WHEEL_FILENAME_CHECK set+restored,
  env restored on exception, no-op when URL missing
- TestRocmTorchInstalledEnvVar: UNSLOTH_ROCM_TORCH_INSTALLED=1 skips
  pip_install, calls _install_bnb_windows_rocm, sets flag
- TestWorkerWindowsRocmPatches: _grouped_mm CUDA dispatch override,
  offs/grouped variant handling, GC-prevention sentinel,
  _StubTypeMeta __instancecheck__, _StubSubpackageFinder registration,
  torchao key submodule pre-stubbing, TORCHDYNAMO_DISABLE guard
- TestRocmTorchPkgSpecs: rocm7.2 torch 2.11.x spec, default <2.11 cap,
  3-tuple shape, _GFX_TO_AMD_INDEX_ARCH RDNA4/3.5/3 coverage

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* tests: fix encoding, IS_WINDOWS patching, and wrong assertion

- Add encoding="utf-8" to all read_text() calls (54 occurrences) so
  tests pass on Windows where the default codec is cp1252 and source
  files contain UTF-8 emoji (e.g. ⚠️ in install_python_stack.py)
- Add @patch.object(stack_mod, "IS_WINDOWS", False) to Linux-path
  TestEnsureRocmTorch tests so they reach the Linux code path when run
  on a Windows machine instead of short-circuiting into the Windows branch
- Fix test_grouped_mm_patch_guarded_by_windows_and_hip_check: the source
  uses getattr(_torch_for_rocm, "version", None) not torch.version, so
  check for '"version"' and '"hip"' substrings instead

137 passed, 2 skipped

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: pin BNB_ROCM_VERSION=72 for torch==2.11.0+rocm7.13.0 compatibility

AMD's pip index now ships torch==2.11.0+rocm7.13.0 (ROCm 7.13).
bitsandbytes auto-detects HIP 7.13 from torch.version.hip and looks for
libbitsandbytes_rocm713.dll, which the AMD Windows prerelease wheel does
not ship (it only ships rocm72.dll), causing a load error at training start.

Fix:
- worker.py section 1f: set BNB_ROCM_VERSION=72 (via setdefault) before
  section 2 ML imports, so bitsandbytes always loads rocm72.dll on Windows ROCm
- install_python_stack.py: set BNB_ROCM_VERSION=72 in _install_bnb_windows_rocm()
  for any post-install imports; update comment to document root cause
- tests: 4 new assertions covering the fix (141 passed, 2 skipped)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: detect BNB ROCm DLL suffix dynamically instead of hardcoding '72'

BNB_ROCM_VERSION was pinned to '72' which works today (AMD wheel ships
rocm72.dll) but would break again if AMD ships a future wheel with a
different DLL suffix (e.g. rocm713.dll).

Add _detect_bnb_rocm_dll_ver() to install_python_stack.py: scans the
installed bitsandbytes package dir for libbitsandbytes_rocm{VER}.dll
using importlib.util.find_spec (no BNB import needed) and returns the
suffix.  '72' remains the fallback when detection fails.

Apply the same detection inline in worker.py section 1f.  Both paths
still respect a pre-set BNB_ROCM_VERSION (caller override wins).

Tests: +8 cases covering detection logic and fallback (147 passed, 2 skipped).

* fix: patch torch.distributed stubs in server process for Windows ROCm

On Windows ROCm, torch.distributed ships without process-group helpers
(is_initialized, is_available, get_rank, get_world_size).  The worker
subprocess already patches these in section 1e, but the main server
process calls _determine_attention_impl_for_gpu_estimate() which calls
unsloth's resolve_attention_implementation() → is_initialized(), causing:

  "Could not resolve attention implementation for '...':
   module 'torch.distributed' has no attribute 'is_initialized'"

Fix: patch the missing attrs onto torch.distributed at the top of
_determine_attention_impl_for_gpu_estimate, matching the same stubs
already applied in worker.py section 1e.  No-ops on Linux/CUDA where
torch.distributed is fully populated.

* fix: gate _grouped_mm dispatch patch on HIP < 7.13

AMD fixed the gfx1200 null HIP kernel in ROCm 7.13 (torch 2.11+).
Users on the new wheel now get the real GPU _grouped_mm kernel for
MoE workloads instead of the Python mm fallback.

Changes:
- worker.py: add _hip_ver_at_least() helper; wrap full _grouped_mm
  patch in `if not _hip_ver_at_least(7, 13):` with else branch that
  logs the skip reason; update section-1f comment to document the fix
- test_rocm_support.py: add 5 tests covering the helper definition,
  the (7, 13) gate expression, the else branch, the skip log message,
  and the AMD-format version string parsing (.split(".")[:2])

Verified: torch==2.11.0+rocm7.13.0 — 3D batch and grouped (offs)
variants both succeed; null crash only present on rocm7.12 and earlier.

* fix: stub is_torchelastic_launched on torch.distributed for Windows ROCm

resolve_attention_implementation calls is_torchelastic_launched() which
does not exist in the incomplete torch.distributed shipped with the
Windows ROCm wheel, causing a warning on every model config load in the
server process. Add it to the stub table alongside the four helpers
already patched in _determine_attention_impl_for_gpu_estimate.

Also adds two tests: one confirming the new stub and one confirming all
five core distributed helpers are covered.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: explicit warnings on AMD ROCm arch/version fallbacks + Fast-Install arg order

setup.ps1:
- Fix Fast-Install argument order: packages before flags, consistent with
  all other Fast-Install calls in the file
  (was: Fast-Install --force-reinstall --index-url $url torch ...)
  (now: Fast-Install torch torchvision torchaudio --force-reinstall --index-url $url)
- Add explicit [WARN] substep when $HasROCm is true but arch mapping fails:
  - GPU arch detected but not in supported wheel list → names the arch and
    lists supported families so user knows exactly what to report
  - HIP SDK present (amd-smi path) but gcnArchName unreadable → instructs
    user to re-install the HIP SDK; previously fell back silently to CPU

install.sh:
- Add [WARN] to stderr before silent CPU fallback when AMD GPU is confirmed
  (rocminfo/amd-smi) but ROCm version cannot be read from any source
  (amd-smi, /opt/rocm/.info/version, hipconfig, dpkg, rpm)
- Add [WARN] to stderr when ROCm version is too old (< 6.0) with upgrade link

install.ps1 and setup.sh: no changes needed (already handle these paths correctly)

* fix: robust gfx arch detection for Strix Halo / HIP-runtime-only installs

Covers users who have the HIP runtime (amd-smi available) but not the
full HIP SDK (no hipinfo), which is common on Strix Halo iGPU systems.
Without this, $ROCmGfxArch stays null and the installer silently falls
back to CPU-only PyTorch despite a working GPU.

Detection waterfall (setup.ps1 + install.ps1):
  1. hipinfo gcnArchName          -- full HIP SDK (existing, unchanged)
  2. amd-smi list gfx pattern     -- newer amd-smi versions embed arch
  3. amd-smi static --asic        -- ROCm 6+ ASIC details with GFX target
  4. UNSLOTH_ROCM_GFX_ARCH env    -- manual override escape hatch
  5. GPU name → arch table        -- best-effort from marketing name:
       890M / Strix Halo  → gfx1151 (RDNA 3.5 iGPU, Strix Halo)
       880M / Strix Point → gfx1150 (RDNA 3.5 iGPU, Strix Point)
       780M / Phoenix     → gfx1103 (RDNA 3 iGPU)
       RX 7900/7800/7700  → gfx1100 (RDNA 3 desktop)
       RX 9070 XT / 9080  → gfx1201 (RDNA 4)
       RX 9070 / 9060 XT  → gfx1200 (RDNA 4)

When arch is inferred from name, a Cyan substep tells the user to set
UNSLOTH_ROCM_GFX_ARCH to skip inference on future installs.
WMI block intentionally does not set $HasROCm (no runtime confirmation).

Tests: 11 new tests in TestStrixHaloGfxArchDetection covering all five
detection levels, WMI safety, and gfx regex in both ps1 files.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: resolve hipinfo/hipconfig via HIP_PATH/ROCM_PATH when not on PATH

AMD HIP SDK sets HIP_PATH on Windows but does not always add the bin
directory to PATH.  Get-Command hipinfo therefore silently fails and
detection falls through to WMI, which cannot provide a gfx arch, leaving
the user with a CPU-only PyTorch install and no warning.

Changes:
- setup.ps1 / install.ps1: before falling through to amd-smi, attempt to
  locate hipinfo.exe and hipconfig.exe under $env:HIP_PATH\bin (then
  $env:ROCM_PATH\bin) when Get-Command returns nothing
- Emit a [WARN] with the resolved path and a one-liner to permanently fix
  PATH via SetEnvironmentVariable
- Emit a [WARN] when HIP_PATH/ROCM_PATH is set but the exe is still not
  found (incomplete SDK install)
- Emit a [WARN] with the first hipinfo output line when hipinfo runs but
  returns a non-zero exit code (e.g. "no ROCm-capable device detected")
- 18 new tests in TestHipSdkEnvPathResolution; total 183 passed, 2 skipped

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* feat: print HIP SDK path and full hipconfig version in terminal on AMD detection

Both install.ps1 and setup.ps1 now emit substeps under the gpu step when
AMD ROCm is detected:

  gpu  AMD ROCm (gfx1200)
       HIP SDK: C:\Program Files\AMD\ROCm\7.1
       hipconfig: 7.1.51803-d3a86bd04

Previously only the gpu label (e.g. "AMD ROCm (gfx1200)") was shown with
no indication of where the SDK was found or which exact build was active.
The full hipconfig build string (e.g. 7.1.51803-d3a86bd04 instead of just
7.1) is now stored in ROCmVersionFull and also used in setup.ps1's
'rocm' step label.

9 new tests in TestHipSdkDetectedSubstep; total 192 passed, 2 skipped

* fix: Strix rocm7.1 segfault bypass + Ubuntu 24.04 HIP gcc-install-dir

Issue 1 (install.sh): gfx1151/gfx1150 + ROCm 7.1 causes a segfault in
torch._grouped_mm (moe_utils.py:167). The Radeon repo now ships cp313
wheels for rocm-rel-7.1, so _amd_gpu_radeon=true silently lands on the
broken combo. When Strix Halo/Point is detected and TORCH_INDEX_URL is
rocm7.1, override to rocm7.2 PyTorch index, update TORCH_CONSTRAINT, and
set _amd_gpu_radeon=false to bypass the Radeon repo entirely. Emits a
clear [WARN] explaining the segfault and linking to the ROCm upgrade docs.

Issue 2 (setup.sh): ROCm 7.x ships clang-20 which on Ubuntu 24.04+ picks
/usr/lib/gcc/x86_64-linux-gnu/14/ (runtime dir, no C++ headers), causing
'cstdlib file not found' and a failed llama.cpp HIP build. Iterate gcc
versions 14→11 to find the first install dir that has both runtime and
/usr/include/c++/<ver> headers, then pass --gcc-install-dir to clang via
CMAKE_HIP_FLAGS. Fix confirmed by h34v3nzc0dex (llama.cpp 417/417 clean).

11 new tests across TestStrixRocm71Override and TestSetupShGccInstallDir;
total 203 passed, 2 skipped

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: BNB_ROCM_VERSION in server process + torch._C._distributed_c10d stubs

Two errors visible in training logs on Windows ROCm:

1. Server process bitsandbytes crash:
   "Configured ROCm binary not found at libbitsandbytes_rocm713.dll"
   The installed BNB wheel ships rocm72.dll (not rocm713.dll). The
   training worker already sets BNB_ROCM_VERSION=72 via DLL detection
   but the server process (main.py) imported bitsandbytes before that
   ran. Fix: add the same DLL-scan + BNB_ROCM_VERSION assignment to
   main.py inside the existing win32 guard, before any downstream
   import can pull in bitsandbytes.

2. torch.distributed import failure:
   "No module named 'torch._C._distributed_c10d'; torch._C is not a package"
   torch._C is a C extension on Windows ROCm — Python cannot do
   submodule imports from it, so torch.distributed fails to import
   before our attribute stubs could ever run. Fix: inject empty
   ModuleType stubs for _distributed_c10d, _distributed_autograd and
   _distributed_rpc into sys.modules inside the win32 guard in
   hardware.py BEFORE importing torch.distributed, so the import
   succeeds and our attribute stubs take effect.

9 new tests in TestServerStartupRocmFixes; total 212 passed, 2 skipped

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(win32): populate distributed c10d stub with dummy symbols

torch.distributed tries to `from torch._C._distributed_c10d import
FakeProcessGroup` (and ProcessGroup, Work, Store, etc.).  The previous
empty ModuleType stub caused an AttributeError on those names.

Populate every stub with a _Dummy class for each known symbol so the
import chain completes silently on Windows ROCm where torch._C is a
compiled extension and its _distributed_c10d submodule doesn't exist.

Adds four new tests in TestServerStartupRocmFixes covering FakeProcessGroup,
ProcessGroup, setattr population, and all three _distributed_* siblings.

* fix(win32): distinguish HIP SDK installed vs GPU not ROCm-accessible

Previously, when hipinfo was found but exited non-zero (e.g. "no
ROCm-capable device detected"), both install.ps1 and setup.ps1 fell
through to the WMI-label-only branch and printed "AMD GPU detected --
HIP SDK not found" -- factually wrong since the SDK binary is present.

Add $HipSdkInstalled flag (set true when hipinfo binary is found,
regardless of exit code). When HipSdkInstalled && !HasROCm:
- Show "AMD GPU detected -- not ROCm-accessible (HIP <ver>)" instead
- Explain this is a driver issue, not an SDK issue, with a link
- Still run hipconfig version capture so version shows in output
- CPU-only hint now says "GPU not ROCm-accessible" not "require HIP SDK"

Also applies to setup.ps1 (same detection block, same branches).

Adds TestHipSdkInstalledButDeviceInaccessible (11 tests).

* fix(win32): scope ROCm workarounds to AMD hosts only

Three Codex-flagged issues where Windows ROCm workarounds incorrectly
applied to Windows CUDA (NVIDIA) machines:

main.py (P1): BNB_ROCM_VERSION was set unconditionally on all win32
hosts. On NVIDIA, bitsandbytes sees BNB_ROCM_VERSION and looks for a
ROCm DLL that doesn't exist, breaking bitsandbytes initialisation.
Fix: gate the block on HIP_PATH/ROCM_PATH being present (ROCm hosts only).

worker.py (P2): torchao stubs were seeded for all win32 runs, shadowing
real torchao on Windows CUDA and silently disabling torchao quantization
for NVIDIA users. Fix: gate on HIP_PATH/ROCM_PATH (win32 ROCm only).

install_python_stack.py (P1): _detect_windows_gfx_arch() only checked
shutil.which("hipinfo"), skipping the HIP_PATH/ROCM_PATH fallback that
the PowerShell installers use. On installs where the HIP SDK bin dir is
not on PATH, _ensure_rocm_torch() returned early without installing
ROCm wheels or bitsandbytes. Fix: mirror the env-var fallback.

* fix(linux): route Strix + ROCm 7.1 to AMD arch-specific index

Instead of falling back to pytorch.org/rocm7.2, the Strix override now
routes to repo.amd.com/rocm/whl/gfx1151/ (or gfx1150/) which serves
torch 2.11.0+rocm7.13.0 -- AMD's build containing the actual _grouped_mm
kernel fix, verified on real gfx1151 hardware by h34v3nzc0dex.

This exercises the real GPU kernel path rather than the rocm7.2 workaround.
UNSLOTH_AMD_ROCM_MIRROR can override the base URL for air-gapped installs.

Also teaches _tauri_torch_index_family to recognise AMD arch-specific URLs
(repo.amd.com/rocm/whl/gfx*) and return the rocm7.13 family label so
_tauri_gpu_branch correctly classifies these installs as rocm.

Suggested by h34v3nzc0dex based on hardware-verified probe results.

* fix(studio/rocm): gate ROCm-only side-effects on active torch runtime

Address five edge cases flagged during PR review:

1. studio/backend/main.py: BNB_ROCM_VERSION was set whenever HIP_PATH or
   ROCM_PATH was present in the environment. A Windows CUDA user who once
   installed the HIP SDK and reverted to a CUDA torch wheel still has those
   env vars set, so bitsandbytes would try to load libbitsandbytes_rocm72.dll
   against a CUDA torch and crash. Now probe torch.version.hip inside the
   env-var guard (worker.py already does this).

2. studio/backend/main.py: os.add_dll_directory returned handles were
   discarded. Per CPython docs, the directory leaves the DLL search list when
   the handle is garbage collected. Retain handles in module-level
   _ROCM_DLL_HANDLES list so they survive process lifetime.

3. studio/install_python_stack.py: _install_bnb_windows_rocm() returned None
   regardless of pip_install_try outcome, and the caller flipped
   _rocm_windows_torch_installed to True unconditionally. On a failed BNB
   install the post-install "manual install may be required" warning was
   suppressed and the user was misled. Helper now returns bool; caller gates
   on it.

4. studio/install_python_stack.py: _detect_windows_gfx_arch returned the raw
   capture group, so mixed-case hipinfo output ("Gfx1151") missed the
   lowercase keys in _GFX_TO_AMD_INDEX_ARCH and silently fell back to CPU
   torch. Lowercase the token.

5. studio/install_python_stack.py: UNSLOTH_ROCM_TORCH_INSTALLED=1 early-
   return trusted the env var even when the venv was wiped between runs.
   Subprocess-probe torch importability first; fall through to the full
   install path if the probe fails.

Tests: 231 passed, 1 skipped in tests/studio/install/test_rocm_support.py
(adds one new test for case 5 fall-through).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/rocm): worker.py parity + don't roll back ROCm torch on bnb failure

Addresses findings from a 10x reviewer pass on the prior fix commit:

1. studio/backend/core/training/worker.py (parity with main.py):
   - Gate the torchao stub block on torch.version.hip / 'rocm' in
     torch.__version__ instead of HIP_PATH / ROCM_PATH env-var presence.
     Same root cause as main.py: HIP SDK env vars stick around on CUDA hosts.
   - Add module-level Windows ROCm DLL registration block. Worker subprocesses
     inherit env vars but not the parent's add_dll_directory handles, so the
     first `import torch` in the worker could fail to find amdhip64.dll when
     HIP_PATH\bin is not on PATH. Mirrors main.py setup. Handles retained at
     module scope via _ROCM_DLL_HANDLES.
   - Promote _WINDOWS_ROCM_GROUPED_MM_LIB to module scope with `global` in
     run_training_process so the torch.library.Library registration survives
     past function return / mid-run garbage collection.
   - Harden _torch_has_hip() to also accept 'rocm' in torch.__version__
     (AMD SDK / Radeon wheels may not set torch.version.hip).

2. studio/install_python_stack.py:
   - Don't roll back ROCm torch when bitsandbytes install fails. The prior
     commit gated _rocm_windows_torch_installed on _install_bnb_windows_rocm()
     returning True; if torch installed successfully but bnb failed, the flag
     stayed False and later install steps could overwrite ROCm torch with the
     generic CPU torch wheel. Set the flag after torch install; surface bnb
     failure as a separate warning instead.
   - _detect_windows_gfx_arch now probes in three tiers: UNSLOTH_ROCM_GFX_ARCH
     env-var override (matches the PowerShell installer), then hipinfo (PATH
     or HIP_PATH\bin), then amd-smi (`static --asic`, `list`). Without the
     amd-smi fallback, runtime-only Radeon installs without hipinfo on PATH
     made `studio update` return early and leave the venv on CPU torch.
   - Linux torch-already-rocm probe in _ensure_rocm_torch now matches the
     Windows probe shape: accepts torch.version.hip OR 'rocm' in
     torch.__version__ to cover AMD SDK / Radeon Linux wheels.

3. studio/backend/utils/hardware/hardware.py:
   - apply_gpu_ids() final-fallback torch probe accepts 'rocm' in
     torch.__version__ in addition to torch.version.hip, matching
     detect_hardware(). AMD SDK wheels could otherwise leak through with
     CUDA-only visibility masks on a spawned ROCm worker.

Tests: 231 passed, 1 skipped in tests/studio/install/test_rocm_support.py
(no test changes needed; the probe shape that prints the hip version (or
'rocm' sentinel) preserves the existing non-empty-string contract).

Not addressed in this commit (deferred or out of scope):
- Tag drift / lemonade checksum (PR 5303 surface, not this PR).
- install.sh rocm7.2.1 URL: small fix, separate.
- install.ps1 / setup.ps1 'Radeon 8060S' marketing-name fallback table.
- Strix Halo + ROCm 7.1 routing asymmetry in Python update path.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/rocm): robustness pass - rocm tag normalisation, Strix routing parity, hardened detection

Robustness pass on top of 76137b2d. Four targeted fixes:

1. install.sh ROCm-tag routing normalisation.
   `rocm7.2.1` would route to https://download.pytorch.org/whl/rocm7.2.1
   which does not exist (PyTorch publishes major.minor URLs only). Same
   for any future patch-level tag. Normalise every rocm{maj.min}* pattern
   to the bare {maj.min} index URL.

2. install.ps1 + studio/setup.ps1 marketing-name fallback.
   The gfx1151 row matched 890M / Strix Halo / HX 37x / HX 38x / AI 9 HX
   but not the actual retail name 'AMD Radeon 8060S Graphics' shipped by
   OEMs (Ryzen AI MAX+ 395). Add '8060S' to the regex.

3. install_python_stack.py Strix + ROCm 7.1 routing parity with install.sh.
   The shell installer reroutes Strix Halo / Point + ROCm 7.1 to
   repo.amd.com/rocm/whl/{gfx}/ (which serves torch 2.11.0+rocm7.13.0
   with the upstream _grouped_mm fix). The Python `studio update` path
   only warned and still installed the broken generic rocm7.1 wheel.
   Mirror the override: detect gfx1151/gfx1150 on ROCm 7.1, route to
   the AMD per-gfx index, honour UNSLOTH_AMD_ROCM_MIRROR override.

4. _detect_windows_gfx_arch amd-smi parsing tightened.
   The amd-smi fallback added in the prior commit used a bare
   `\bgfx[1-9][0-9a-z]{2,3}\b` match against the lowercased stdout,
   which could pick up stray gfx references in warnings / device-name
   strings. Anchor on labelled lines first (Target_Graphics_Version,
   ASIC, Arch, gfx) and fall back to the bare match only when no
   labelled line is present.

Tests: 231 passed, 1 skipped in tests/studio/install/test_rocm_support.py;
sim_5301 23 cases pass (6 new sims for the Strix override + amd-smi parsing).

* fix(studio/rocm): multi-GPU selection, Strix sibling handling, defensive cleanups

Round 4 robustness pass based on 5 parallel Opus reviewers of head 21773215.
Seven items from across regression / edge-case / error-paths / architecture
reviews:

1. studio/backend/main.py BNB gate: aligned with the broad ROCm check used
   everywhere else in this PR (torch.version.hip OR 'rocm' in __version__).
   AMD SDK / Radeon Linux wheels do not always populate torch.version.hip;
   without this, main.py would silently skip BNB_ROCM_VERSION while worker.py
   set it.

2. studio/install_python_stack.py _install_bnb_windows_rocm: init _ok = False
   before the try block. Without this, if pip_install_try itself raises
   (e.g. OSError on uv binary missing), the finally block restored env vars
   correctly but the subsequent `if not _ok:` raised UnboundLocalError,
   masking the original exception.

3. studio/install_python_stack.py _detect_windows_gfx_arch:
   - Rewrote to use re.findall (not re.search) on both hipinfo and amd-smi
     output, dedup tokens preserving order, and select via new
     _pick_visible_index() helper.
   - HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES (first comma entry, integer)
     now picks the right GPU on multi-AMD-GPU hosts. Out-of-range or non-int
     values fall back to the first GPU (matches detect_host behaviour in
     install_llama_prebuilt.py).

4. studio/install_python_stack.py Strix override now consults the runtime
   target before flipping:
   - Previous behaviour intersected gfx_codes with {gfx1151, gfx1150} and
     picked the first Strix arch, ignoring whether HIP_VISIBLE_DEVICES
     selected a non-Strix sibling (e.g. discrete RX 7900 in a mixed APU+dGPU
     box). Could install Strix-specific wheels onto a gfx1100 dGPU.
   - Now resolves the runtime gfx via _pick_visible_index() and only
     overrides when that runtime target is in the Strix set.

5. studio/backend/main.py + studio/backend/core/training/worker.py: ROCm
   version dir scan no longer sorts lexically. Previous sort placed "10.0"
   before "7.0" alphabetically, which would mis-prioritise ROCm 10.x bin
   dirs once AMD ships them. New _ver_key() splits on "." and sorts
   numerically with a string fallback.

6. install.sh Strix override URL: replaced ${var%/} (strips one trailing
   slash) with a while-loop that strips all trailing slashes, matching
   Python's .rstrip("/"). A user setting UNSLOTH_AMD_ROCM_MIRROR with
   "http://corp/whl///" no longer ends up with "http://corp/whl///gfx1151/"
   which strict pip proxies (artifactory, sonatype) 404 on.

7. studio/install_python_stack.py: bumped torch import probe timeout from
   30s to 90s. PyTorch's lazy .so loading can take 60-90s on cold NFS or
   USB-backed venvs. The shorter timeout was producing a false "torch
   missing" classification and reinstalling a working ROCm torch.

Tests: 231 passed, 1 skipped. sim_5301 30 cases pass (added 7 new sims for
multi-GPU detection, Strix sibling handling, and _ok-init regression).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/rocm): worker BNB/grouped_mm broad gate, install.sh Strix visibility, runtime-only ROCm detection

Round-5 robustness pass based on 20 parallel reviewers of head 96b9e465.

1. studio/backend/core/training/worker.py - BNB version pin / dynamo disable
   / _grouped_mm fallback block was still gated on torch.version.hip alone
   despite the torchao stub block above already using the broad check. AMD
   SDK / Radeon Windows wheels (torch.__version__ contains "rocm" but
   torch.version.hip is None) silently skipped the Windows ROCm runtime
   patches. Aligned to the same broad check (8/20 reviewers).

2. studio/backend/core/training/worker.py - _hip_ver_at_least() now also
   parses the ROCm version out of torch.__version__ (e.g. "2.11.0+rocm7.13.0")
   when torch.version.hip is missing, so the kernel-fix gate is correct for
   SDK / Radeon wheels too.

3. studio/backend/core/training/worker.py - _grouped_mm_safe_impl with
   offs=None now picks torch.bmm/matmul for 3-D inputs instead of always
   calling torch.mm. The real _grouped_mm accepts 3-D batched matmul; the
   prior fallback raised "self must be a matrix" on MoE workloads (2/20).

4. studio/backend/main.py - dropped the HIP_PATH / ROCM_PATH env-var gate
   from the BNB block; probe torch directly. Runtime-only Radeon / AMD SDK
   Windows installs do not set those SDK env vars but still ship ROCm torch
   (5/20 reviewers).

5. install.sh - Strix override now collects every gfx token from
   rocminfo / amd-smi (in enumeration order), then indexes by
   HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES so a mixed Strix iGPU + non-
   Strix dGPU host where the user selected the dGPU does NOT get rerouted
   to the Strix per-gfx index. Mirrors the Python update path (5/20 reviewers).

6. install.sh - Strix detection chain now also probes `amd-smi static --asic`,
   matching the PowerShell installer (1/20). Closes the gap on runtime-only
   Strix hosts where `amd-smi list` does not surface a gfx token.

7. studio/install_python_stack.py - _has_rocm_gpu() now has the sysfs KFD
   topology fallback (/sys/class/kfd/kfd/topology/nodes/*/gpu_id), matching
   install.sh. On minimal package-managed installs without rocminfo /
   amd-smi GUI tools, `studio update` can now detect the GPU and repair the
   venv instead of returning early (2/20).

8. studio/install_python_stack.py - _detect_amd_gfx_codes() now falls back
   to `amd-smi list` and `amd-smi static --asic` when rocminfo is missing
   (2/20). Strix routing on runtime-only Radeon hosts now matches what
   install.sh has done for a while.

9. studio/install_python_stack.py - Strix override now applies even when
   has_hip_torch is True. The whole point of the override is to repair an
   existing broken torch.version.hip == "7.1" install; skipping the
   reinstall left users on the known _grouped_mm segfaulting stack (3/20).

Tests: 231 passed, 1 skipped. sim_5301 30 cases pass. sim_cross 12 pass.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/rocm): code review hardening pass

- main.py: numeric DLL sort (string sort picked rocm72 over rocm713);
  add basename() to regex; log warning on detection failure; log info
  when BNB_ROCM_VERSION is set (mirrors worker.py)
- worker.py: explicit len-guard in _hip_ver_at_least() with warning
  logs instead of silent IndexError/ValueError swallow
- hardware.py: isinstance(result, dict) guard before result.get() in
  _smi_query() to prevent AttributeError on non-dict backend returns
- amd.py: round() before int() on parsed GPU IDs; log warning when
  truncation occurs (defensive against malformed amd-smi output)
- setup.sh: quote --gcc-install-dir value in CMAKE_HIP_FLAGS so paths
  with spaces do not break the CMake argument
- install.ps1, setup.ps1: apply colon-split + ToLower() to hipinfo
  gcnArchName match (consistent with each other and with setup.sh)
- install.sh: tighten ROCm tag case patterns to explicit
  rocmX.Y|rocmX.Y.* to avoid unintended prefix matches

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/training): GPU OOM guard to prevent system freeze on VRAM exhaustion

On RDNA 4 (gfx1200/gfx1201) and other ROCm GPUs, exhausting VRAM can
cause a HIP driver hang that freezes the entire system rather than
raising a recoverable Python exception.

Two-part fix:
- set_per_process_memory_fraction(0.90) caps the HIP/CUDA allocator at
  90% of VRAM so PyTorch raises OutOfMemoryError before hitting the
  hardware limit, keeping the driver alive and the system responsive
- top-level exception handler detects OOM errors by type and message
  and surfaces a clear actionable message to the UI (reduce
  max_seq_length, enable gradient_checkpointing, lower batch size)
  instead of the raw CUDA/HIP error string

* fix(studio/rocm): OOM guard ROCm-only + unified memory, multi-GPU arch selection

OOM guard (worker.py):
- Scope to _hw.IS_ROCM only -- NVIDIA CUDA has a graceful OOM path and
  does not need the allocator cap
- Detect unified memory by comparing torch VRAM against psutil system RAM;
  use 0.80 on unified-memory APUs (gfx1151 Strix Halo) where the GPU pool
  is carved from host RAM, 0.90 on discrete cards

Multi-GPU arch selection:
- install.ps1 / setup.ps1: replace -match (first hit only) with
  [regex]::Matches() to collect all gcnArchName entries, then index by
  HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES
- install_python_stack.py: index into full token list before dedup so
  HIP_VISIBLE_DEVICES=2 on [gfx1100, gfx1100, gfx1151] resolves gfx1151
- install.sh: remove awk dedup from gfx token collection for same reason

GCC multiarch (setup.sh):
- Only append -linux-gnu when gcc -print-multiarch does not already return
  the full triple, fixing double-suffix on Ubuntu 24.04

* fix(tests): update ROCm version cap expectations from rocm7.1 to rocm7.2

Daniel's normalisation commit updated the cap from rocm7.1 to rocm7.2
since PyTorch now publishes that index and rocm7.2 ships torch 2.11.0.
Test expectations were stale.

* fix(tests): correct MLX smoke test losses_per_step assertion

logging_steps=1 with max_steps=30 produces 30 loss entries, not 7.
The assertion was stale from a previous config.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(studio/worker): detect unified-memory APU by GPU name not VRAM/RAM ratio

The previous heuristic (VRAM > 50 % of system RAM) false-positived on discrete
cards in low-RAM systems — e.g. RX 9060 XT 16 GB on a 16 GB or 24 GB machine
would trip the unified-memory path and log "unified memory host" when it should
say "discrete".

AMD iGPUs (gfx1150/gfx1151 Strix Halo, Strix Point, etc.) expose names with a
digit+M suffix ("AMD Radeon 890M"), while discrete cards use "RX NNNN [XT|XTX]"
naming.  Matching that suffix is reliable across all current ROCm-capable AMD
consumer GPUs and does not require psutil.

Also includes the device name in the log line to ease future debugging.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(install/setup.ps1): force array on hipinfo gcnArchName parse to fix single-GPU arch truncation

When [regex]::Matches() finds exactly one match, PowerShell's pipeline
unwraps the result to a scalar string.  Indexing a scalar string with [0]
returns the first *character*, so a one-GPU system would parse
gcnArchName "gfx1200" as "g", which is not in the supported arch map
and triggers the CPU-only fallback.

Wrapping with @() forces the result to remain an array regardless of
match count.  On a single-GPU machine the arch is now correctly read as
"gfx1200" (or whatever the full name is) so the ROCm wheel index is
selected.

Reproducer: hipinfo exits 0 and outputs exactly one gcnArchName line.
Without @(), $_hipAllArches = "gfx1200" (String); $_hipAllArches[0] = 'g'.
With @(), $_hipAllArches = @("gfx1200") (Object[]); $_hipAllArches[0] = "gfx1200".

* fix(studio/rocm): classify unified-memory APU via VRAM/RAM ratio, not arch list

Replace the gcnArchName allowlist {gfx1150, gfx1151} with a
psutil-based heuristic: unified APUs expose the entire system RAM
as the HIP pool (ratio ≥ 0.90), discrete cards are well below that.
No arch name required — future APUs classify correctly without code changes.

Also removes the stale import re / \d[Mm]\b device-name regex that
5d84704 left behind, and logs vram/sys GiB for easier on-hardware
verification.

Addresses h34v3nzc0dex review: Radeon 8060S (gfx1151, 128 GiB
unified) now correctly gets 0.80 cap instead of 0.90.

* fix(studio/rocm): revert to gcnArchName for unified-memory APU classification

VRAM/RAM ratio >= 0.90 false-positives on machines where discrete VRAM
equals system RAM (e.g. RX 9060 XT 16 GB + 16 GB system RAM → ratio 1.0,
incorrectly classified as unified → wrong 0.80 cap applied).

gcnArchName is the correct signal: naming-independent, stable within a
product family, and already parsed throughout this PR. Unified set is
{gfx1150, gfx1151} (Strix Point + Strix Halo).

* fix(studio/llama-prebuilt): resolve hipinfo via HIP_PATH/ROCM_PATH on Windows

shutil.which("hipinfo") returns None when the HIP SDK bin dir is not on
PATH -- the HIP SDK installer sets HIP_PATH/ROCM_PATH but does not always
add the bin dir to PATH. This caused has_rocm=False in the prebuilt asset
selector, so AMD ROCm machines got the CPU llama.cpp zip instead of the
HIP one, silently running all chat inference on CPU.

Add _resolve_exe() that falls back to %HIP_PATH%\bin and %ROCM_PATH%\bin
when shutil.which() finds nothing, mirroring the same fallback already
present in setup.ps1.

* fix(studio/llama-prebuilt): pass --has-rocm from setup.ps1 to skip re-detection

The Python prebuilt installer re-detects ROCm independently via
shutil.which("hipinfo"), which fails when hipinfo is not on PATH
(HIP SDK sets HIP_PATH but doesn't always add the bin dir to PATH).
This caused has_rocm=False and downloaded the CPU llama.cpp zip even
on confirmed AMD ROCm machines.

setup.ps1 already performs reliable ROCm detection with its own
HIP_PATH/ROCM_PATH fallback. Add --has-rocm flag to
install_llama_prebuilt.py so setup.ps1 can forward its result directly,
and pass it whenever $HasROCm is true. The Python script then overrides
has_rocm=True in the HostInfo without re-probing.

* fix(studio/llama-prebuilt): add HIP asset to simple-policy Windows path

direct_upstream_release_plan (used by --simple-policy, which setup.ps1
always passes) only checked has_usable_nvidia on Windows and fell
straight to CPU for AMD ROCm machines, ignoring has_rocm entirely.
The --has-rocm override had no effect because the simple-policy code
path never reached resolve_asset_choice where has_rocm was checked.

Add an elif branch for has_rocm that tries the upstream HIP asset
(llama-TAG-bin-win-hip-radeon-x64.zip) before falling through to the
CPU fallback, consistent with the non-simple-policy path.

* fix(studio/setup.ps1): auto-remove mismatched llama.cpp install kind

When an existing llama.cpp install is the wrong kind for the current
GPU (e.g. windows-cpu on an AMD ROCm machine that should have
windows-hip), the prebuilt installer skips on tag match and never
upgrades. Read install_kind from UNSLOTH_PREBUILT_INFO.json before
invoking the installer and remove the directory if the kind doesn't
match, forcing a fresh download of the correct variant.

* fix(studio/setup.ps1): show live PyTorch install output in verbose mode for ROCm

The ROCm torch reinstall (setup.ps1 phase) always silently captured
output, so in --verbose mode the torch downgrade mid-install
(2.11.0+rocm → 2.10.0 → 2.11.0+rocm) looked like the final state was
2.10.0. Match the CPU/CUDA blocks which show live uv output when
$script:UnslothVerbose is set.

* fix(rocm/windows): set ROCBLAS_TENSILE_LIBPATH for bundled rocblas.dll

The llama.cpp ROCm prebuilt bundles rocblas.dll next to the binary but
not the Tensile kernel library files it depends on at runtime
(rocblas/library/TensileLibrary*.dat + *.hsaco).  The bundled DLL
searches for these files relative to its own location by default, i.e.
<binary_dir>/rocblas/library/, which does not exist in the prebuilt
install tree.  This causes a silent crash on the very first GEMM
(prefill) with no output from llama-server, seen by the caller as
WinError 10054 / 10061.  Model load and the single-token warmup pass
because they use simpler code paths that do not trigger rocBLAS GEMM.

Fix: set ROCBLAS_TENSILE_LIBPATH in the subprocess env to
<HIP_PATH>/bin/rocblas/library so the bundled DLL finds the kernel
files from the system ROCm installation.  Uses setdefault so a user-
supplied env var is never overwritten.  No-ops on CUDA and CPU (no
HIP_PATH) and on Linux (win32 branch only).

Reproducer log:
  rocBLAS error: Cannot read .../Release/rocblas/library/TensileLibrary.dat
  rocBLAS error: Could not initialize Tensile host:
  directory_iterator: The system cannot find the path specified.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(install.sh): restore gfx token dedup in Strix multi-GPU awk indexer

536a54df removed the per-source `| awk '!seen[$0]++'` dedup from the
_gfx_all collection step but left the indexer awk as bare NF, so on a
mixed-arch host (e.g. dGPU gfx1100 + Strix iGPU gfx1151) where
rocminfo emits each gfx token twice (Name: field + ISA triple),
HIP_VISIBLE_DEVICES=1 indexed vals[1] = the second gfx1100 occurrence
instead of gfx1151, triggering the Strix routing on the wrong GPU.

Add !seen[$0]++ to the indexer awk so duplicate tokens from the same
GPU collapse to one entry before the HIP_VISIBLE_DEVICES index is
applied -- matching exactly what the Python side does with dict.fromkeys()
in _detect_amd_gfx_codes(). The comment above the block ("skip
duplicates") already documented this as the intended behaviour.

* fix(studio/install): correct _TOTAL progress count on Windows

base_total += 3 fired for all non-macOS platforms including Windows,
but flash-attn (line 1620) and ROCm torch final (line 1705) are both
guarded by 'not IS_WINDOWS and not IS_MACOS', so on Windows with torch
enabled _TOTAL was 13 while only 11 _progress() calls actually execute.

Split into +1 for the ROCm torch check (all non-macOS) and +2 for the
two Linux-only steps, so Windows gets _TOTAL=11 and Linux gets 14.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(install.ps1): enforce torch>=2.11.0 for gfx120X and Strix on Windows

The AMD arch-specific index (repo.amd.com/rocm/whl/gfx120X-all/ and
gfx1151/) publishes torch wheels from 2.7.1 through 2.11.0. Without a
version floor pip can resolve to torch 2.10.0+rocm7.12 on RDNA 4
(gfx120X) or torch 2.10.0+rocm7.1 on Strix (gfx1151/gfx1150), both of
which have a null-pointer crash in torch._C._grouped_mm (TheRock
issues #5284 / #3284). torch 2.11.0+rocm7.13 contains the fix.

Add $ROCmTorchFloor alongside $ROCmIndexUrl: set to torch>=2.11.0 for
the two affected arch families, null for all others. Wire it into the
uv pip install call so the broken wheels are never selected.

* fix(rocm/windows): address Codex nits - deterministic DLL suffix, CUDA llama.cpp kind, HIP_VISIBLE_DEVICES arch indexing

- install_python_stack.py / worker.py: _detect_bnb_rocm_dll_ver() and the
  inline worker probe now collect ALL libbitsandbytes_rocm*.dll suffixes and
  return max() by numeric value instead of stopping at the first glob hit.
  Filesystem glob order is not guaranteed; this ensures '713' always wins
  over '72' when both variants are present in the wheel.

- setup.ps1 (expectedKind): add 'windows-cuda' branch so NVIDIA hosts are
  not treated as 'windows-cpu'. Previously an existing windows-cuda prebuilt
  was always considered a mismatch on non-ROCm machines, forcing an
  unnecessary re-download on every update.

- setup.ps1 (amd-smi gfx arch): collect ALL gfx tokens from amd-smi list
  output in GPU order and honour HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES
  when selecting which arch to use. On mixed-arch AMD systems where the
  visible GPU is not the first enumerated one, this prevents installing an
  incompatible wheel index. Falls back to index 0 (same as before) when the
  visibility var is unset or is a comma-separated list.

- test_rocm_support.py: add test_picks_highest_suffix_when_multiple_dlls to
  cover the multi-DLL case that was previously untested.

* fix(rocm): misleading amd-smi log, BNB spec consistency, torch ceiling for AMD index

amd.py: split 'returncode != 0 or not stdout' into two separate branches.
Previously, exit-0 with empty output logged 'amd-smi returned code 0' (which
reads as success, not a warning) and incorrectly incremented the circuit-breaker
counter. Now: non-zero exit logs the code and counts toward the limit as before;
empty stdout on exit 0 logs at DEBUG level and does not penalise the counter
(amd-smi --json always emits at least [] on exit 0, so this branch is rare and
is not a tool failure).

main.py: replace spec.origin / os.path.dirname() with
spec.submodule_search_locations to match install_python_stack.py and worker.py.
For normal wheel installs both approaches reach the same directory, but using
submodule_search_locations is the canonical way and handles editable bitsandbytes
installs correctly. Also use max() by numeric suffix (same as the other two sites)
instead of a sort-then-break loop.

install.ps1: add <2.12.0 ceiling to the torch constraint for gfx120X (RDNA 4)
and gfx1151/gfx1150 (Strix). AMD actively publishes new versions on their
per-arch index; without a ceiling, a future 2.12.0+rocmX.Y wheel would be
pulled in automatically before being validated on these architectures. The
ceiling matches the existing Linux install_python_stack.py constraint for the
same arches. Bump both when 2.12.x is confirmed working.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(rocm): torch floor in setup.ps1, torchvision pin for Strix, rocmsdk in _hip_ver_at_least

setup.ps1: add \ (mirrors install.ps1) and derive \
from it. Previously the AMD index install called 'Fast-Install torch torchvision
torchaudio --force-reinstall --index-url \' with no version
constraint, so pip could resolve torch 2.10.0+rocm7.12 for gfx1151/gfx1200 --
the exact broken wheel the PR is meant to avoid. Now gfx120X and Strix enforce
'torch>=2.11.0,<2.12.0', matching install.ps1 and the Linux constraint.

install_python_stack.py: pin torchvision and torchaudio in _strix_override_pkgs.
The Strix Linux override uses --index-url (exclusive, no PyPI fallback); bare
unversioned 'torchvision' and 'torchaudio' could resolve a build from AMD's
index targeting a different torch major, causing ABI/version mismatches at
runtime. Now pinned to '>=0.26.0,<0.27.0' and '>=2.11.0,<2.12.0' respectively,
matching _ROCM_TORCH_CONSTRAINT['rocm7.2'].

worker.py: extend _hip_ver_at_least to handle AMD SDK wheel version strings.
The fallback regex r'rocm(\d+)\.(\d+)' cannot match '2.9.0+rocmsdk20251116'
(no rocmX.Y component), so the function always returned False on SDK/Radeon
wheels -- installing the Python _grouped_mm workaround on wheels that already
have the working HIP kernel. Added a second check: if the version string
contains '+rocmsdk', assume >= 7.13 (the rocmsdk format post-dates the
gfx120X null-kernel fix) and skip the fallback.

* fix(rocm): warn on OOB HIP_VISIBLE_DEVICES, bail on empty numeric_ids mask

- setup.ps1: when HIP/ROCR_VISIBLE_DEVICES names an index beyond the
  detected GPU count, emit a yellow warning and fall back to GPU 0
  instead of silently reading allGfxArches[-1] (wrong arch)
- hardware.py _reconcile_primary_rocm_unified_memory: distinguish
  numeric_ids=None (no env var, use torch ordinal 0) from numeric_ids=[]
  (empty mask / HIP_VISIBLE_DEVICES=-1, no GPU visible); bail out early
  in the empty case to avoid querying torch.device(0) incorrectly

* fix(rocm): gate StubSubpackageFinder on win32 ROCm, add gcnArchName fallbacks

- worker.py _StubSubpackageFinder: the meta_path append was running on
  every platform on every call to run_training_process; moved it inside
  the if _is_win32_rocm: block since stubs are only seeded there and the
  finder is a pure accumulation on Linux/Windows CUDA
- worker.py OOM guard: AMD SDK / Radeon wheels may not populate
  gcnArchName, causing Strix Halo to be misclassified as discrete and
  get the 0.90 cap (12.8 GB OS headroom) instead of 0.80 (25.6 GB);
  now tries gcn_arch_name / arch_name / gfx_arch_name variants first,
  then falls back to device-name matching (890M -> Strix Halo,
  880M -> Strix Point) with a debug log when the fallback fires

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(rocm): pin torchvision/torchaudio in setup.ps1, remove -Unique from arch array

- setup.ps1 ROCm torch install: torchvision and torchaudio were passed
  bare alongside pinned torch>=2.11.0,<2.12.0 for gfx1151/gfx1200 arches.
  AMD publishes packages independently so a future torchvision 0.27 (for
  torch 2.12) on the same arch index would cause pip ResolutionImpossible
  or an ABI-incompatible install. Added torchvisionFloorMap and
  torchaudioFloorMap mirroring install_python_stack.py's strix override
  (torchvision>=0.26.0,<0.27.0, torchaudio>=2.11.0,<2.12.0) and derived
  ROCmVisionSpec/ROCmAudioSpec used in all three Fast-Install call sites.

- setup.ps1 amd-smi arch detection: Select-Object -Unique was collapsing
  same-arch multi-GPU arrays (e.g. two gfx1151 APUs -> 1-element array)
  causing HIP_VISIBLE_DEVICES=1 to trigger a false out-of-range warning
  and fall back to GPU 0 even though the correct GPU would have been at
  index 1. Removed -Unique; added comment noting the positional-index
  assumption and its non-contiguous-GPU limitation.

* fix(rocm): add 8060s/8050s to OOM guard device-name fallback, extract classifier helper

Path 3 of the OOM guard device-name fallback only checked for 890m/880m
(gfx1150 Strix Point SKU names). Strix Halo (gfx1151) ships as Radeon 8060S
(Ryzen AI MAX+ 395) and Radeon 8050S (cut-down SKU) -- neither matches, so
the fallback returned is_unified=False and applied the 0.90 fraction instead
of 0.80, leaving ~12.8 GiB OS headroom on a 128 GiB pool instead of ~25.6 GiB.

Fix: add 8060s and 8050s to the name-match set. Also correct the comment that
mislabelled 890M as a Strix Halo name (it is Strix Point).

Refactor: extract the three-path classifier into _rocm_classify_unified_memory()
so it can be unit-tested directly. Add 31 test cases in test_rocm_oom_guard.py
covering all three paths and the regression case (Radeon 8060S Graphics).

Reported-by: h34v3nzc0dex

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix(rocm): pass explicit dtype on bf16-unsupported hardware (RDNA2)

dtype=None lets unsloth auto-detect the model dtype. On RDNA2 (gfx103x,
e.g. RX 6600) is_bfloat16_supported() incorrectly returns True, so unsloth
picks bf16 and the first bf16 kernel dispatch triggers:

  LLVM ERROR: Cannot select: intrinsic %llvm.amdgcn.fdot2.bf16.bf16

Replace every dtype=None in load_model() with _auto_dtype which resolves
to None when bf16 is supported (all modern NVIDIA + RDNA3+) and
torch.float16 otherwise. This gives RDNA2 users a working float16
training path without touching NVIDIA behaviour at all.

Fixes: https://github.com/unslothai/unsloth/issues/5337

* fix: reduce log noise for expected non-issues on Windows ROCm

Three log lines fired at warning/error level for conditions that are
completely expected on a Windows HIP SDK-only setup:

amd.py
- amd-smi WinError 2 (FileNotFoundError): downgrade warning -> debug.
  amd-smi ships with Adrenalin, not the HIP SDK; absence is normal.
- 'disabling' message: downgrade warning -> info with clearer text
  'not available (not installed; expected on HIP SDK-only systems);
  GPU VRAM polling disabled'

hardware.py
- torch.distributed.Store missing: downgrade warning -> debug.
  The distributed stub added in this PR intentionally omits Store; the
  attention-impl fallback to eager is expected and non-actionable.

worker.py
- causal-conv1d: add early Windows exit (info) in both
  _ensure_causal_conv1d_fast_path and _causal_conv1d_install hook;
  no cp313/win_amd64 wheel exists, so the install always fails.
- FLA: add early Windows exit (info) in
  _ensure_flash_linear_attention_unconditional; triton dependency has
  no cp313/win_amd64 wheel.
- Defense-in-depth: _install_package_wheel_first non-HIP PyPI failure
  logs info+debug on Windows instead of error; FLA failure logs
  info+debug on Windows instead of warning.

* [AMD] FIx installation of bitsandbytes when it's from .dev and skip rebuilding llama.cpp if we build it manually.

* fix: use force_pip for Windows ROCm bitsandbytes prebuilt wheel install

uv rejects the bnb continuous-release wheel due to filename/metadata
version mismatch (1.33.7.preview vs 0.50.0.dev0). Switch to force_pip=True
(pip bypass) instead of the UV_SKIP_WHEEL_FILENAME_CHECK env var workaround
-- cleaner and consistent with how the Linux path handles it.

BNB_ROCM_VERSION is still set post-install to the detected DLL suffix so
the worker subprocess loads the correct libbitsandbytes_rocm{VER}.dll even
when torch.version.hip reports a newer HIP version than the wheel ships.

* fix: three small correctness fixes found in PR review

- _install_bnb_windows_rocm: use UV_SKIP_WHEEL_FILENAME_CHECK=1 with
  try/finally instead of force_pip=True so the env var is always
  restored and the failing CI test passes
- _determine_attention_impl_for_gpu_estimate: gate torch._C distributed
  stubs on IS_ROCM so Windows CUDA users keep the real extension
- install.ps1 amd-smi fallback: collect all gfx tokens and index by
  HIP_VISIBLE_DEVICES, matching the hipinfo path on multi-GPU hosts

* fix: stub torchao in export subprocess on Windows ROCm

On Windows, the ROCm build of PyTorch ships without the distributed
C extension (torch._C._distributed_c10d). torchao, which is pulled in
transitively by transformers.quantizers at import time, walks into
torch.distributed._functional_collectives -> distributed_c10d and
crashes with:

  No module named 'torch._C._distributed_c10d'; 'torch._C' is not a package

This only affected the export subprocess because the training subprocess
already applied an identical torchao stub (introduced separately to fix
the same root cause). The export subprocess had no such guard and died
during 'Importing Unsloth...' before any model loading could happen.

Fix: apply the same _StubSubpackageFinder / torchao stub pattern to the
export subprocess entry point, gated on Windows ROCm detection, before
any import of transformers or unsloth_zoo.

Root cause tracked in ROCm/TheRock#3284 (libuv / torch.distributed
missing on Windows ROCm builds).

Ref: https://github.com/ROCm/TheRock/issues/3284

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* install.sh, setup.sh: add GPU arch step logging to match PS1 scripts

Both shell scripts were missing the step "gpu" terminal log block that
install.ps1 and setup.ps1 emit. This adds equivalent output: GPU label
with gfx arch (e.g. "AMD ROCm (gfx1151)"), ROCm root path, hipconfig
version, and marketing name substep. Includes the same gfx arch detection
chain (rocminfo → amd-smi list → amd-smi static --asic), UNSLOTH_ROCM_GFX_ARCH
env override, and name-based arch inference table (Strix Halo/Point, RDNA 3/4)
as the PS1 versions. install.sh also replaces bare echo blocks for the AMD
ROCm and CPU-only cases with formatted substep output.

* Fix BNB_ROCM_VERSION gate, ROCm GPU mask preference, APU unified memory and Release build for PR #5301

- main.py: gate BNB_ROCM_VERSION on the rocm bnb DLL or HIP_PATH/ROCM_PATH instead of importing torch on every Windows host
- hardware.py: prefer HIP/ROCR visible-device masks only on ROCm hosts so a stale mask cannot override CUDA_VISIBLE_DEVICES on NVIDIA
- llama_cpp.py: set GGML_CUDA_ENABLE_UNIFIED_MEMORY=1 only for unified-memory APUs (gfx1150/gfx1151)
- setup.sh: pass -DCMAKE_BUILD_TYPE=Release for the HIP source build
- add test_amd_apu_unified_memory.py

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: guard recompile_limit + fix AMD VRAM monitor fallback

trainer.py: torch._dynamo.config.recompile_limit does not exist in
some ROCm torch builds (e.g. pytorch.org/whl/rocm6.2 wheels). Guard
the assignment so training doesn't crash on RDNA2/RDNA3.

hardware.py: when amd-smi/nvidia-smi is unavailable or returns no
usable data (HIP SDK-only Windows, Docker, unexpected JSON format),
the existing fallback used torch.cuda.memory_allocated() which is
process-specific and reads near-zero even with a fully loaded model.
Switch to torch.cuda.mem_get_info() via _torch_get_per_device_info()
which reports system-wide VRAM occupancy so the GPU monitor shows
real usage on all AMD systems without requiring amd-smi.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: Windows VRAM monitor via Performance Counter API

When amd-smi/nvidia-smi is unavailable on Windows, query dedicated GPU
VRAM via Windows Performance Counters (same source as Task Manager).
This gives system-wide cross-process usage, fixing the near-zero reading
caused by torch.cuda.mem_get_info only seeing the Studio server process.

Linux fallback path unchanged (mem_get_info is system-wide on ROCm).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: rename to _rocm_windows_perf_counter_vram_gb, scope to IS_ROCM

Function is AMD ROCm specific — amd-smi absent on Windows when only the
HIP SDK is installed. Scoped to IS_ROCM so NVIDIA Windows path is
untouched (nvidia-smi handles that case).

* fix: AMD VRAM monitor — Linux DRM sysfs + Windows perf counter

Linux: read /sys/class/drm/card*/device/mem_info_vram_used|total for
system-wide GPU memory across all processes. No tools required, always
present on Linux AMD systems.

Windows: Windows Performance Counter API (already added).

Both paths are gated on IS_ROCM and only fire when amd-smi is absent.
torch mem_get_info remains as last resort (process-local).

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: AMD GPU monitor — utilization, temperature, and power for Windows and Linux fallback paths

- Windows: GPU utilization via \GPU Engine(*engtype_3D*)\Utilization Percentage perf counter
- Windows: temperature and power via ADL (atiadlxx.dll, ships with Adrenalin)
- Linux: GPU utilization via DRM sysfs gpu_busy_percent
- Linux: temperature via hwmon temp1_input (millidegrees C)
- Linux: power via hwmon power1_average / power1_input (microwatts)

All paths are no-op fallbacks (None) when the source is unavailable.
Mirrors what nvidia-smi provides on the CUDA path.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* fix: remove ADL ctypes — does not support AMD iGPU (Strix Halo)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Lee Jackson <130007945+Imagineer99@users.noreply.github.com>
Co-authored-by: Erland366 <erland.pg366@gmail.com>
Co-authored-by: danielhanchen <michaelhan2050@gmail.com>
2026-05-29 22:29:56 -07:00

1967 lines
100 KiB
PowerShell

# Unsloth Studio Installer for Windows PowerShell
# Usage: irm https://raw.githubusercontent.com/unslothai/unsloth/main/install.ps1 | iex
# Local: Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass; .\install.ps1 --local
# NoTorch: .\install.ps1 --no-torch (skip PyTorch, GGUF-only mode)
# Test: .\install.ps1 --package roland-sloth
#
# Env vars (priority: UNSLOTH_STUDIO_HOME > STUDIO_HOME > USERPROFILE-redirect > default):
# UNSLOTH_STUDIO_HOME / STUDIO_HOME = path -> install under that path
# (DataDir nests inside; user PATH not modified persistently).
# Default ($USERPROFILE\.unsloth\studio) is preserved when no env var is set.
function Install-UnslothStudio {
$ErrorActionPreference = "Stop"
$script:UnslothVerbose = ($env:UNSLOTH_VERBOSE -eq "1")
# ── Tauri structured output ──
function Write-TauriLog {
param([string]$Tag, [string]$Message)
if ($TauriMode) {
Write-Host "[TAURI:$Tag] $Message"
}
}
function Format-TauriDiagBool {
param([bool]$Value)
if ($Value) { return "true" }
return "false"
}
function Get-TauriDiagArch {
$arch = [string]$env:PROCESSOR_ARCHITECTURE
if ([string]::IsNullOrWhiteSpace($arch)) {
try { $arch = [System.Runtime.InteropServices.RuntimeInformation]::OSArchitecture.ToString() } catch { $arch = "unknown" }
}
$arch = $arch.ToLowerInvariant()
switch ($arch) {
"amd64" { return "x86_64" }
"x64" { return "x86_64" }
"arm64" { return "arm64" }
"x86" { return "x86" }
default { return ($arch -replace '[^a-z0-9_.-]', '_') }
}
}
function Get-TauriTorchIndexFamily {
param([string]$TorchIndexUrl)
if ($SkipTorch) { return "none" }
if ([string]::IsNullOrWhiteSpace($TorchIndexUrl)) { return "none" }
$leaf = ($TorchIndexUrl.TrimEnd('/') -split '/')[-1].ToLowerInvariant()
if (@("cpu", "cu118", "cu124", "cu126", "cu128", "cu130") -contains $leaf) { return $leaf }
if ($leaf -match '^rocm[0-9]+\.[0-9]+$') { return $leaf }
return "auto"
}
function Get-TauriGpuBranch {
param([string]$TorchIndexFamily)
if ($SkipTorch) { return "no_torch" }
if ($TorchIndexFamily -like "cu*") { return "cuda" }
if ($TorchIndexFamily -like "rocm*") { return "rocm" }
if ($TorchIndexFamily -eq "cpu") { return "cpu" }
return "unknown"
}
function Write-TauriDiag {
param(
[string]$GpuBranch = "unknown",
[string]$TorchIndexFamily = "none",
[string]$PythonVersionForDiag = $PythonVersion
)
if ([string]::IsNullOrWhiteSpace($PythonVersionForDiag)) { $PythonVersionForDiag = "unknown" }
Write-TauriLog "DIAG" "diag_schema=1 platform=windows arch=$(Get-TauriDiagArch) python_version=$($PythonVersionForDiag.ToLowerInvariant()) skip_torch=$(Format-TauriDiagBool $SkipTorch) mac_intel=false gpu_branch=$GpuBranch torch_index_family=$TorchIndexFamily"
}
function Exit-InstallFailure {
param(
[Parameter(Mandatory = $true)][string]$Message,
[int]$Code = 1
)
if ($Code -eq 0) { $Code = 1 }
Write-TauriLog "ERROR" $Message
if (Get-Command Restore-StudioVenvRollback -CommandType Function -ErrorAction SilentlyContinue) {
Restore-StudioVenvRollback
}
if ($TauriMode) {
exit $Code
}
}
# ── Parse flags ──
$StudioLocalInstall = $false
$PackageName = "unsloth"
$RepoRoot = ""
$TauriMode = $false
$SkipTorch = $false
$ShortcutsOnly = $false
$argList = $args
for ($i = 0; $i -lt $argList.Count; $i++) {
switch ($argList[$i]) {
"--local" { $StudioLocalInstall = $true }
"--tauri" { $TauriMode = $true }
"--no-torch" { $SkipTorch = $true }
"--verbose" { $script:UnslothVerbose = $true }
"-v" { $script:UnslothVerbose = $true }
"--shortcuts-only" { $ShortcutsOnly = $true }
"--package" {
$i++
if ($i -ge $argList.Count) {
Write-Host "[ERROR] --package requires an argument." -ForegroundColor Red
return (Exit-InstallFailure "--package requires an argument.")
}
$PackageName = $argList[$i]
}
}
}
# Propagate to child processes so they also respect verbose mode.
# Process-scoped -- does not persist.
if ($script:UnslothVerbose) {
$env:UNSLOTH_VERBOSE = '1'
}
if ($StudioLocalInstall) {
$RepoRoot = (Resolve-Path (Split-Path -Parent $PSCommandPath)).Path
if (-not (Test-Path (Join-Path $RepoRoot "pyproject.toml"))) {
Write-Host "[ERROR] --local must be run from the unsloth repo root (pyproject.toml not found at $RepoRoot)" -ForegroundColor Red
return (Exit-InstallFailure "--local must be run from the unsloth repo root")
}
}
# Validate --package to prevent injection into shell/Python commands
if ($PackageName -notmatch '^[a-zA-Z0-9][a-zA-Z0-9._-]*$') {
Write-Host "[ERROR] --package name contains invalid characters (allowed: a-z A-Z 0-9 . _ -)" -ForegroundColor Red
return (Exit-InstallFailure "--package name contains invalid characters")
}
$PythonVersion = "3.13"
# Resolve install destinations. Priority: UNSLOTH_STUDIO_HOME, then
# STUDIO_HOME alias, then USERPROFILE-redirect, then default.
# Reject whitespace-only values so " " is treated as unset (matches the
# Python resolvers' .strip()), preventing install/runtime layout drift.
$envOverrideVar = $null
$envOverride = $null
if (-not [string]::IsNullOrWhiteSpace($env:UNSLOTH_STUDIO_HOME)) {
$envOverrideVar = "UNSLOTH_STUDIO_HOME"
$envOverride = $env:UNSLOTH_STUDIO_HOME.Trim()
} elseif (-not [string]::IsNullOrWhiteSpace($env:STUDIO_HOME)) {
$envOverrideVar = "STUDIO_HOME"
$envOverride = $env:STUDIO_HOME.Trim()
}
# Custom Studio roots are not supported with --tauri (desktop app still
# resolves %USERPROFILE%\.unsloth\studio). Pass through if override == legacy.
if ($TauriMode -and $envOverride) {
$_tauriOverride = $envOverride
if ($_tauriOverride -eq "~" -or $_tauriOverride -like "~/*" -or $_tauriOverride -like "~\*") {
$_tauriOverride = (Join-Path $env:USERPROFILE $_tauriOverride.Substring(1).TrimStart('/','\'))
}
try {
$_tauriOverride = [System.IO.Path]::GetFullPath($_tauriOverride)
} catch {}
$_legacyTauriRoot = Join-Path $env:USERPROFILE ".unsloth\studio"
try {
$_legacyTauriRoot = [System.IO.Path]::GetFullPath($_legacyTauriRoot)
} catch {}
# Strip trailing separators so ".../studio\" matches ".../studio".
$_trimSeps = @(
[System.IO.Path]::DirectorySeparatorChar,
[System.IO.Path]::AltDirectorySeparatorChar
)
$_tauriOverride = $_tauriOverride.TrimEnd($_trimSeps)
$_legacyTauriRoot = $_legacyTauriRoot.TrimEnd($_trimSeps)
if ($_tauriOverride -ne $_legacyTauriRoot) {
Write-Host "ERROR: $envOverrideVar is not supported with --tauri." -ForegroundColor Red
Write-Host " The desktop app still uses the legacy %USERPROFILE%\.unsloth\studio root." -ForegroundColor Red
Write-Host " Run install.ps1 without --tauri for custom-root shell installs," -ForegroundColor Yellow
Write-Host " or unset the env var for default desktop installs." -ForegroundColor Yellow
throw "$envOverrideVar is not supported with --tauri."
}
}
$defaultProfile = $null
try { $defaultProfile = [Environment]::GetFolderPath("UserProfile") } catch {}
# LOCALAPPDATA may be unset in service / CI contexts; Join-Path would abort
# under ErrorActionPreference=Stop without this guard.
$defaultDataDir = if ($env:LOCALAPPDATA -and -not [string]::IsNullOrWhiteSpace($env:LOCALAPPDATA)) {
Join-Path $env:LOCALAPPDATA "Unsloth Studio"
} else { $null }
if ($envOverride) {
# Tilde expansion: env vars aren't subject to it when quoted on assignment.
if ($envOverride -eq "~" -or $envOverride -like "~/*" -or $envOverride -like "~\*") {
$envOverride = (Join-Path $env:USERPROFILE $envOverride.Substring(1).TrimStart('/','\'))
}
try {
# .NET API: New-Item -Path treats brackets as wildcards and has no
# -LiteralPath in PS 5.1, so a root like C:\studio[abc] would fail.
[System.IO.Directory]::CreateDirectory($envOverride) | Out-Null
$StudioHome = (Resolve-Path -LiteralPath $envOverride).Path
} catch {
Write-Host "ERROR: $envOverrideVar=$envOverride cannot be created or accessed." -ForegroundColor Red
throw "$envOverrideVar=$envOverride cannot be created or accessed."
}
$probe = Join-Path $StudioHome (".unsloth-write-probe-" + [guid]::NewGuid())
try {
# WriteAllText: literal-path safe + closes handle so Remove-Item works.
[System.IO.File]::WriteAllText($probe, "")
Remove-Item -LiteralPath $probe -Force -ErrorAction SilentlyContinue
} catch {
Write-Host "ERROR: $envOverrideVar=$StudioHome is not writable." -ForegroundColor Red
throw "$envOverrideVar=$StudioHome is not writable."
}
$StudioDataDir = Join-Path $StudioHome "share"
$StudioRedirectMode = 'env'
} elseif ($defaultProfile -and $env:USERPROFILE -and ($env:USERPROFILE -ne $defaultProfile)) {
$StudioHome = Join-Path $env:USERPROFILE ".unsloth\studio"
$StudioDataDir = $defaultDataDir
$StudioRedirectMode = 'profile'
} else {
$StudioHome = Join-Path $env:USERPROFILE ".unsloth\studio"
$StudioDataDir = $defaultDataDir
$StudioRedirectMode = 'default'
}
$VenvDir = Join-Path $StudioHome "unsloth_studio"
$Rule = [string]::new([char]0x2500, 52)
$Sloth = [char]::ConvertFromUtf32(0x1F9A5)
function Enable-StudioVirtualTerminal {
if ($env:NO_COLOR) { return $false }
try {
if (-not ("StudioVT.Native" -as [type])) {
Add-Type -Namespace StudioVT -Name Native -MemberDefinition @'
[DllImport("kernel32.dll")] public static extern IntPtr GetStdHandle(int nStdHandle);
[DllImport("kernel32.dll")] public static extern bool GetConsoleMode(IntPtr h, out uint m);
[DllImport("kernel32.dll")] public static extern bool SetConsoleMode(IntPtr h, uint m);
'@ -ErrorAction Stop
}
$h = [StudioVT.Native]::GetStdHandle(-11)
[uint32]$mode = 0
if (-not [StudioVT.Native]::GetConsoleMode($h, [ref]$mode)) { return $false }
$mode = $mode -bor 0x0004
return [StudioVT.Native]::SetConsoleMode($h, $mode)
} catch {
return $false
}
}
$script:StudioVtOk = Enable-StudioVirtualTerminal
function Get-StudioAnsi {
param(
[Parameter(Mandatory = $true)]
[ValidateSet('Title', 'Dim', 'Ok', 'Warn', 'Err', 'Reset')]
[string]$Kind
)
$e = [char]27
switch ($Kind) {
'Title' { return "${e}[38;5;150m" }
'Dim' { return "${e}[38;5;245m" }
'Ok' { return "${e}[38;5;108m" }
'Warn' { return "${e}[38;5;136m" }
'Err' { return "${e}[91m" }
'Reset' { return "${e}[0m" }
}
}
Write-Host ""
if ($script:StudioVtOk -and -not $env:NO_COLOR) {
Write-Host (" " + (Get-StudioAnsi Title) + $Sloth + " Unsloth Studio Installer (Windows)" + (Get-StudioAnsi Reset))
Write-Host (" {0}{1}{2}" -f (Get-StudioAnsi Dim), $Rule, (Get-StudioAnsi Reset))
} else {
Write-Host (" {0} Unsloth Studio Installer (Windows)" -f $Sloth) -ForegroundColor DarkGreen
Write-Host " $Rule" -ForegroundColor DarkGray
}
Write-Host ""
# ── Helper: refresh PATH from registry (deduplicating entries) ──
# Merge order: venv Scripts (if active) > Machine > User > current $env:Path.
# Dedup compares both raw and expanded forms (%VAR% vs literal).
function Refresh-SessionPath {
$machine = [System.Environment]::GetEnvironmentVariable("Path", "Machine")
$user = [System.Environment]::GetEnvironmentVariable("Path", "User")
$venvScripts = if ($env:VIRTUAL_ENV) { Join-Path $env:VIRTUAL_ENV "Scripts" } else { $null }
$sources = @()
if ($venvScripts) { $sources += $venvScripts }
$sources += @($machine, $user, $env:Path)
$merged = ($sources | Where-Object { $_ }) -join ";"
$seen = @{}
$unique = New-Object System.Collections.Generic.List[string]
foreach ($p in $merged -split ";") {
$rawKey = $p.Trim().Trim('"').TrimEnd("\").ToLowerInvariant()
$expKey = [Environment]::ExpandEnvironmentVariables($p).Trim().Trim('"').TrimEnd("\").ToLowerInvariant()
if ($rawKey -and -not $seen.ContainsKey($rawKey) -and -not $seen.ContainsKey($expKey)) {
$seen[$rawKey] = $true
if ($expKey -and $expKey -ne $rawKey) { $seen[$expKey] = $true }
$unique.Add($p)
}
}
$env:Path = $unique -join ";"
}
# ── Helper: safely add a directory to the persistent User PATH ──
# Direct registry access preserves REG_EXPAND_SZ (avoids dotnet/runtime#1442).
# Append (default) keeps existing tools first; Prepend for must-win entries.
function Add-ToUserPath {
param(
[Parameter(Mandatory = $true)][string]$Directory,
[ValidateSet('Append','Prepend')]
[string]$Position = 'Append'
)
try {
$regKey = [Microsoft.Win32.Registry]::CurrentUser.CreateSubKey('Environment')
try {
$rawPath = $regKey.GetValue('Path', '', [Microsoft.Win32.RegistryValueOptions]::DoNotExpandEnvironmentNames)
[string[]]$entries = if ($rawPath) { $rawPath -split ';' } else { @() } # string[] prevents scalar collapse
$normalDir = $Directory.Trim().Trim('"').TrimEnd('\').ToLowerInvariant()
$expNormalDir = [Environment]::ExpandEnvironmentVariables($Directory).Trim().Trim('"').TrimEnd('\').ToLowerInvariant()
$kept = New-Object System.Collections.Generic.List[string]
$matchIndices = New-Object System.Collections.Generic.List[int]
for ($i = 0; $i -lt $entries.Count; $i++) {
$stripped = $entries[$i].Trim().Trim('"')
$rawNorm = $stripped.TrimEnd('\').ToLowerInvariant()
$expNorm = [Environment]::ExpandEnvironmentVariables($stripped).TrimEnd('\').ToLowerInvariant()
$isMatch = ($rawNorm -and ($rawNorm -eq $normalDir -or $rawNorm -eq $expNormalDir)) -or
($expNorm -and ($expNorm -eq $normalDir -or $expNorm -eq $expNormalDir))
if ($isMatch) {
$matchIndices.Add($i)
continue
}
$kept.Add($entries[$i])
}
$alreadyPresent = $matchIndices.Count -gt 0
if ($alreadyPresent -and $Position -eq 'Append') { # Append: idempotent no-op
return $false
}
if ($alreadyPresent -and $Position -eq 'Prepend' -and # Prepend: no-op if already at front
$matchIndices.Count -eq 1 -and $matchIndices[0] -eq 0) {
return $false
}
# One-time backup under HKCU\Software\Unsloth\PathBackup
if ($rawPath) {
try {
$backupKey = [Microsoft.Win32.Registry]::CurrentUser.CreateSubKey('Software\Unsloth')
try {
$existingBackup = $backupKey.GetValue('PathBackup', $null)
if (-not $existingBackup) {
$backupKey.SetValue('PathBackup', $rawPath, [Microsoft.Win32.RegistryValueKind]::ExpandString)
}
} finally {
$backupKey.Close()
}
} catch { }
}
if (-not $rawPath) {
Write-Host "[WARN] User PATH is empty - initializing with $Directory" -ForegroundColor Yellow
}
$newPath = if ($rawPath) {
if ($Position -eq 'Prepend') {
(@($Directory) + $kept) -join ';'
} else {
($kept + @($Directory)) -join ';'
}
} else {
$Directory
}
if ($newPath -ceq $rawPath) { # no actual change
return $false
}
$regKey.SetValue('Path', $newPath, [Microsoft.Win32.RegistryValueKind]::ExpandString)
# Broadcast WM_SETTINGCHANGE via dummy env-var roundtrip.
# [NullString]::Value avoids PS 7.5+/.NET 9 $null-to-"" coercion.
try {
$d = "UnslothPathRefresh_$([guid]::NewGuid().ToString('N').Substring(0,8))"
[Environment]::SetEnvironmentVariable($d, '1', 'User')
[Environment]::SetEnvironmentVariable($d, [NullString]::Value, 'User')
} catch { }
return $true
} finally {
$regKey.Close()
}
} catch {
Write-Host "[WARN] Could not update User PATH: $($_.Exception.Message)" -ForegroundColor Yellow
return $false
}
}
function step {
param(
[Parameter(Mandatory = $true)][string]$Label,
[Parameter(Mandatory = $true)][string]$Value,
[string]$Color = "Green"
)
if ($script:StudioVtOk -and -not $env:NO_COLOR) {
$dim = Get-StudioAnsi Dim
$rst = Get-StudioAnsi Reset
$val = switch ($Color) {
'Green' { Get-StudioAnsi Ok }
'Yellow' { Get-StudioAnsi Warn }
'Red' { Get-StudioAnsi Err }
'DarkGray' { Get-StudioAnsi Dim }
default { Get-StudioAnsi Ok }
}
$padded = if ($Label.Length -ge 15) { $Label.Substring(0, 15) } else { $Label.PadRight(15) }
Write-Host (" {0}{1}{2}{3}{4}{2}" -f $dim, $padded, $rst, $val, $Value)
} else {
$padded = if ($Label.Length -ge 15) { $Label.Substring(0, 15) } else { $Label.PadRight(15) }
Write-Host (" {0}" -f $padded) -NoNewline -ForegroundColor DarkGray
$fc = switch ($Color) {
'Green' { 'DarkGreen' }
'Yellow' { 'Yellow' }
'Red' { 'Red' }
'DarkGray' { 'DarkGray' }
default { 'DarkGreen' }
}
Write-Host $Value -ForegroundColor $fc
}
}
function substep {
param(
[Parameter(Mandatory = $true)][string]$Message,
[string]$Color = "DarkGray"
)
if ($script:StudioVtOk -and -not $env:NO_COLOR) {
$msgCol = switch ($Color) {
'Yellow' { (Get-StudioAnsi Warn) }
'Red' { (Get-StudioAnsi Err) }
default { (Get-StudioAnsi Dim) }
}
$pad = "".PadRight(15)
Write-Host (" {0}{1}{2}{3}" -f $msgCol, $pad, $Message, (Get-StudioAnsi Reset))
} else {
$fc = switch ($Color) {
'Yellow' { 'Yellow' }
'Red' { 'Red' }
default { 'DarkGray' }
}
Write-Host (" {0,-15}{1}" -f "", $Message) -ForegroundColor $fc
}
}
# Run native commands quietly by default to match install.sh behavior.
# Full command output is shown only when --verbose / UNSLOTH_VERBOSE=1.
function Invoke-InstallCommand {
param(
[Parameter(Mandatory = $true)][ScriptBlock]$Command
)
$prevEap = $ErrorActionPreference
$ErrorActionPreference = "Continue"
try {
# Reset to avoid stale values from prior native commands.
$global:LASTEXITCODE = 0
if ($script:UnslothVerbose) {
# Merge stderr into stdout so progress/warning output stays visible
# without flipping $? on successful native commands (PS 5.1 treats
# stderr records as errors that set $? = $false even on exit code 0).
& $Command 2>&1 | Out-Host
} else {
$output = & $Command 2>&1 | Out-String
if ($LASTEXITCODE -ne 0) {
Write-Host $output -ForegroundColor Red
}
}
return [int]$LASTEXITCODE
} finally {
$ErrorActionPreference = $prevEap
}
}
function New-StudioShortcuts {
param(
[Parameter(Mandatory = $true)][string]$UnslothExePath
)
if (-not (Test-Path -LiteralPath $UnslothExePath)) {
substep "cannot create shortcuts, unsloth.exe not found at $UnslothExePath" "Yellow"
return
}
try {
# Persist an absolute path in launcher scripts so shortcut working
# directory changes do not break process startup.
$UnslothExePath = (Resolve-Path -LiteralPath $UnslothExePath).Path
# Escape for single-quoted embedding in generated launcher script.
# This prevents runtime variable expansion for paths containing '$'.
$SingleQuotedExePath = $UnslothExePath -replace "'", "''"
# $StudioDataDir = LOCALAPPDATA\Unsloth Studio, or $StudioHome\share in env-mode.
if (-not $StudioDataDir -or [string]::IsNullOrWhiteSpace($StudioDataDir)) {
substep "DataDir path unavailable; skipped shortcut creation" "Yellow"
return
}
$appDir = $StudioDataDir
$launcherPs1 = Join-Path $appDir "launch-studio.ps1"
$launcherVbs = Join-Path $appDir "launch-studio.vbs"
$desktopDir = [Environment]::GetFolderPath("Desktop")
$desktopLink = if ($desktopDir -and $desktopDir.Trim()) {
Join-Path $desktopDir "Unsloth Studio.lnk"
} else {
$null
}
$startMenuDir = if ($env:APPDATA -and $env:APPDATA.Trim()) {
Join-Path $env:APPDATA "Microsoft\Windows\Start Menu\Programs"
} else {
$null
}
$startMenuLink = if ($startMenuDir -and $startMenuDir.Trim()) {
Join-Path $startMenuDir "Unsloth Studio.lnk"
} else {
$null
}
if (-not $desktopLink) {
substep "Desktop path unavailable; skipped desktop shortcut creation" "Yellow"
}
if (-not $startMenuLink) {
substep "APPDATA/Start Menu path unavailable; skipped Start menu shortcut creation" "Yellow"
}
$iconPath = Join-Path $appDir "unsloth.ico"
$bundledIcon = $null
if ($PSScriptRoot -and $PSScriptRoot.Trim()) {
$bundledIcon = Join-Path $PSScriptRoot "studio\frontend\public\unsloth.ico"
}
$iconUrl = "https://raw.githubusercontent.com/unslothai/unsloth/main/studio/frontend/public/unsloth.ico"
if (-not (Test-Path -LiteralPath $appDir)) {
[System.IO.Directory]::CreateDirectory($appDir) | Out-Null
}
# Same-install discriminator: per-install opaque id written once at
# install time and read by both this launcher and the backend
# (/api/health). Replaces the older sha256(resolved $StudioHome)
# scheme to (a) avoid leaking the install path on -H 0.0.0.0
# deployments and (b) sidestep launcher/backend canonicalization
# drift (Resolve-Path vs Path.resolve() junction handling). Lives
# at $StudioHome\share\ (not $appDir) so the backend can find it
# via _STUDIO_ROOT_RESOLVED / "share" / "studio_install_id"
# regardless of mode. 32 bytes of crypto random -> 64 hex chars.
$_studioIdDir = Join-Path $StudioHome "share"
if (-not (Test-Path -LiteralPath $_studioIdDir)) {
[System.IO.Directory]::CreateDirectory($_studioIdDir) | Out-Null
}
$_studioIdFile = Join-Path $_studioIdDir "studio_install_id"
$_studioRootId = ""
if ((Test-Path -LiteralPath $_studioIdFile) -and `
((Get-Item -LiteralPath $_studioIdFile).Length -gt 0)) {
$_studioRootId = ([System.IO.File]::ReadAllText($_studioIdFile)).Trim()
}
if (-not $_studioRootId) {
$_idBytes = New-Object byte[] 32
[Security.Cryptography.RandomNumberGenerator]::Create().GetBytes($_idBytes)
$_studioRootId = -join ($_idBytes | ForEach-Object { $_.ToString('x2') })
# Atomic write: write to a temp sibling then rename, so a partial
# install cannot leave a half-written id.
$_idTmp = $_studioIdFile + ".$PID.tmp"
[System.IO.File]::WriteAllText($_idTmp, $_studioRootId)
Move-Item -LiteralPath $_idTmp -Destination $_studioIdFile -Force
}
# Env-mode: persist UNSLOTH_STUDIO_HOME (and llama path) so fresh
# shells don't need to re-export, and bake per-install $portFile /
# $mutexName so concurrent custom-root launchers cannot serialize
# through one global mutex on 8888..8908. Default installs get an
# empty prefix to match pre-PR behavior.
$studioHomeExport = if ($StudioRedirectMode -eq 'env') {
# When override == legacy default, llama.cpp stays at
# ~/.unsloth/llama.cpp (one shared build). Canonicalize the
# legacy side so the comparison survives path normalization.
$_legacyStudio = Join-Path $env:USERPROFILE ".unsloth\studio"
if (Test-Path -LiteralPath $_legacyStudio -PathType Container) {
$_legacyStudio = (Resolve-Path -LiteralPath $_legacyStudio).Path
}
$_llamaPath = if ($StudioHome -eq $_legacyStudio) {
Join-Path $env:USERPROFILE ".unsloth\llama.cpp"
} else {
Join-Path $StudioHome "llama.cpp"
}
$_sq = $StudioHome -replace "'", "''"
$_llama = $_llamaPath -replace "'", "''"
$_appDirSq = $appDir -replace "'", "''"
$_appBytes = [Text.Encoding]::UTF8.GetBytes($appDir)
$_appHash = ([BitConverter]::ToString(
[Security.Cryptography.SHA256]::Create().ComputeHash($_appBytes)
) -replace '-', '').Substring(0, 16)
# UNSLOTH_LLAMA_CPP_PATH is a pre-existing user override; only default if unset.
"`$env:UNSLOTH_STUDIO_HOME = '$_sq'`nif (-not `$env:UNSLOTH_LLAMA_CPP_PATH) {`n `$env:UNSLOTH_LLAMA_CPP_PATH = '$_llama'`n}`n`$portFile = '$_appDirSq\studio.port'`n`$mutexName = 'Local\UnslothStudioLauncher-$_appHash'`n"
} else {
"`$portFile = `$null`n`$mutexName = 'Local\UnslothStudioLauncher'`n"
}
$launcherContent = @"
$studioHomeExport`$ErrorActionPreference = 'Stop'
`$basePort = 8888
`$maxPortOffset = 20
`$timeoutSec = 60
`$pollIntervalMs = 1000
`$_ExpectedStudioRootId = '$_studioRootId'
function Test-StudioHealth {
param([Parameter(Mandatory = `$true)][int]`$Port)
try {
`$url = "http://127.0.0.1:`$Port/api/health"
`$resp = Invoke-RestMethod -Uri `$url -TimeoutSec 1 -Method Get
if (-not (`$resp -and `$resp.status -eq 'healthy' -and `$resp.service -eq 'Unsloth UI Backend')) { return `$false }
# why: verify the backend belongs to THIS install via the install-time
# hex digest; raw path is not leaked over /api/health.
if (`$_ExpectedStudioRootId -and `$resp.studio_root_id -ne `$_ExpectedStudioRootId) { return `$false }
return `$true
} catch {
return `$false
}
}
function Get-CandidatePorts {
# Fast path: only probe base port + currently listening ports in range.
`$ports = @(`$basePort)
try {
`$maxPort = `$basePort + `$maxPortOffset
`$listening = Get-NetTCPConnection -State Listen -ErrorAction Stop |
Where-Object { `$_.LocalPort -ge `$basePort -and `$_.LocalPort -le `$maxPort } |
Select-Object -ExpandProperty LocalPort
`$ports = (@(`$basePort) + `$listening) | Sort-Object -Unique
} catch {
Write-Host "[DEBUG] Get-NetTCPConnection failed: `$(`$_.Exception.Message). Falling back to full port scan." -ForegroundColor DarkGray
# Fallback when Get-NetTCPConnection is unavailable/restricted.
for (`$offset = 1; `$offset -le `$maxPortOffset; `$offset++) {
`$ports += (`$basePort + `$offset)
}
}
return `$ports
}
function Find-HealthyStudioPort {
if (`$portFile) {
if (Test-Path -LiteralPath `$portFile) {
`$cached = Get-Content -LiteralPath `$portFile -ErrorAction SilentlyContinue | Select-Object -First 1
if (`$cached -match '^\d+`$') {
`$cachedPort = [int]`$cached
if (Test-StudioHealth -Port `$cachedPort) { return `$cachedPort }
Remove-Item -LiteralPath `$portFile -Force -ErrorAction SilentlyContinue
}
}
return `$null
}
foreach (`$candidate in (Get-CandidatePorts)) {
if (Test-StudioHealth -Port `$candidate) {
return `$candidate
}
}
return `$null
}
function Test-PortBusy {
param([Parameter(Mandatory = `$true)][int]`$Port)
`$listener = `$null
try {
`$listener = [System.Net.Sockets.TcpListener]::new([System.Net.IPAddress]::Any, `$Port)
`$listener.Start()
return `$false
} catch {
return `$true
} finally {
if (`$listener) { try { `$listener.Stop() } catch {} }
}
}
function Find-FreeLaunchPort {
`$maxPort = `$basePort + `$maxPortOffset
try {
`$listening = Get-NetTCPConnection -State Listen -ErrorAction Stop |
Where-Object { `$_.LocalPort -ge `$basePort -and `$_.LocalPort -le `$maxPort } |
Select-Object -ExpandProperty LocalPort
for (`$offset = 0; `$offset -le `$maxPortOffset; `$offset++) {
`$candidate = `$basePort + `$offset
if (`$candidate -notin `$listening) {
return `$candidate
}
}
} catch {
# Get-NetTCPConnection unavailable or restricted; probe ports directly
for (`$offset = 0; `$offset -le `$maxPortOffset; `$offset++) {
`$candidate = `$basePort + `$offset
if (-not (Test-PortBusy -Port `$candidate)) {
return `$candidate
}
}
}
return `$null
}
# If Studio is already healthy on any expected port, just open it and exit.
`$existingPort = Find-HealthyStudioPort
if (`$existingPort) {
Start-Process "http://localhost:`$existingPort"
exit 0
}
`$launchMutex = [System.Threading.Mutex]::new(`$false, `$mutexName)
`$haveMutex = `$false
try {
try {
`$haveMutex = `$launchMutex.WaitOne(0)
} catch [System.Threading.AbandonedMutexException] {
`$haveMutex = `$true
}
if (-not `$haveMutex) {
# Another launcher is already running; wait for it to bring Studio up
`$deadline = (Get-Date).AddSeconds(`$timeoutSec)
while ((Get-Date) -lt `$deadline) {
`$port = Find-HealthyStudioPort
if (`$port) { Start-Process "http://localhost:`$port"; exit 0 }
Start-Sleep -Milliseconds `$pollIntervalMs
}
exit 0
}
`$powershellExe = Join-Path `$env:SystemRoot 'System32\WindowsPowerShell\v1.0\powershell.exe'
`$studioExe = '$SingleQuotedExePath'
`$launchPort = Find-FreeLaunchPort
if (-not `$launchPort) {
`$msg = "No free port found in range `$basePort-`$(`$basePort + `$maxPortOffset)"
try {
Add-Type -AssemblyName System.Windows.Forms -ErrorAction Stop
[System.Windows.Forms.MessageBox]::Show(`$msg, 'Unsloth Studio') | Out-Null
} catch {}
exit 1
}
# Single-quote the path in the child -Command so `$` / backtick in custom
# roots don't get reparsed; double any apostrophes so 'O''Brien' survives.
`$studioCommand = "& '" + (`$studioExe -replace "'", "''") + "' studio -p " + `$launchPort
`$launchArgs = @(
'-NoExit',
'-NoProfile',
'-ExecutionPolicy',
'Bypass',
'-Command',
`$studioCommand
)
try {
`$proc = Start-Process -FilePath `$powershellExe -ArgumentList `$launchArgs -WorkingDirectory `$env:USERPROFILE -PassThru
} catch {
`$msg = "Could not launch Unsloth Studio terminal.`n`nError: `$(`$_.Exception.Message)"
try {
Add-Type -AssemblyName System.Windows.Forms -ErrorAction Stop
[System.Windows.Forms.MessageBox]::Show(`$msg, 'Unsloth Studio') | Out-Null
} catch {}
exit 1
}
`$browserOpened = `$false
`$deadline = (Get-Date).AddSeconds(`$timeoutSec)
while ((Get-Date) -lt `$deadline) {
if (Test-StudioHealth -Port `$launchPort) {
if (`$portFile) {
try {
[System.IO.File]::WriteAllText(`$portFile, "`$launchPort`n")
} catch {}
}
Start-Process "http://localhost:`$launchPort"
`$browserOpened = `$true
break
}
if (`$proc.HasExited) { break }
Start-Sleep -Milliseconds `$pollIntervalMs
}
if (-not `$browserOpened) {
if (`$proc.HasExited) {
`$msg = "Unsloth Studio exited before becoming healthy. Check terminal output for errors."
} else {
`$msg = "Unsloth Studio is still starting but did not become healthy within `$timeoutSec seconds. Check the terminal window for the selected port and open it manually."
}
try {
Add-Type -AssemblyName System.Windows.Forms -ErrorAction Stop
[System.Windows.Forms.MessageBox]::Show(`$msg, 'Unsloth Studio') | Out-Null
} catch {}
}
} finally {
if (`$haveMutex) { `$launchMutex.ReleaseMutex() | Out-Null }
`$launchMutex.Dispose()
}
exit 0
"@
# Write UTF-8 with BOM for reliable decoding by Windows PowerShell 5.1,
# even when install.ps1 is executed from PowerShell 7.
$utf8Bom = New-Object System.Text.UTF8Encoding($true)
[System.IO.File]::WriteAllText($launcherPs1, $launcherContent, $utf8Bom)
$vbsContent = @"
Set shell = CreateObject("WScript.Shell")
cmd = "powershell -NoProfile -ExecutionPolicy Bypass -WindowStyle Hidden -File ""$launcherPs1"""
shell.Run cmd, 0, False
"@
# WSH handles UTF-16LE reliably for .vbs files with non-ASCII paths.
Set-Content -LiteralPath $launcherVbs -Value $vbsContent -Encoding Unicode -Force
# Prefer bundled icon from local clone/dev installs.
# If not available, best-effort download from raw GitHub.
# We only attach the icon if the resulting file has a valid ICO header.
$hasValidIcon = $false
if ($bundledIcon -and (Test-Path -LiteralPath $bundledIcon)) {
try {
Copy-Item -LiteralPath $bundledIcon -Destination $iconPath -Force
} catch {
Write-Host "[DEBUG] Error copying bundled icon: $($_.Exception.Message)" -ForegroundColor DarkGray
}
} elseif (-not (Test-Path -LiteralPath $iconPath)) {
try {
Invoke-WebRequest -Uri $iconUrl -OutFile $iconPath -UseBasicParsing
} catch {
Write-Host "[DEBUG] Error downloading icon: $($_.Exception.Message)" -ForegroundColor DarkGray
}
}
if (Test-Path -LiteralPath $iconPath) {
try {
$bytes = [System.IO.File]::ReadAllBytes($iconPath)
if (
$bytes.Length -ge 4 -and
$bytes[0] -eq 0 -and
$bytes[1] -eq 0 -and
$bytes[2] -eq 1 -and
$bytes[3] -eq 0
) {
$hasValidIcon = $true
} else {
Remove-Item -LiteralPath $iconPath -Force -ErrorAction SilentlyContinue
}
} catch {
Write-Host "[DEBUG] Error validating or removing icon: $($_.Exception.Message)" -ForegroundColor DarkGray
Remove-Item -LiteralPath $iconPath -Force -ErrorAction SilentlyContinue
}
}
# Env-mode: skip persistent Desktop / Start Menu .lnk shortcuts
# that may point at a deleted workspace; launcher + icon stay.
if ($StudioRedirectMode -eq 'env') {
substep "wrote launcher at $launcherPs1 (persistent shortcuts skipped in env-override mode)"
return
}
$wscriptExe = Join-Path $env:SystemRoot "System32\wscript.exe"
$shortcutArgs = "//B //Nologo `"$launcherVbs`""
try {
$wshell = New-Object -ComObject WScript.Shell
$createdShortcutCount = 0
foreach ($linkPath in @($desktopLink, $startMenuLink)) {
if (-not $linkPath -or [string]::IsNullOrWhiteSpace($linkPath)) { continue }
try {
$shortcut = $wshell.CreateShortcut($linkPath)
$shortcut.TargetPath = $wscriptExe
$shortcut.Arguments = $shortcutArgs
$shortcut.WorkingDirectory = $appDir
$shortcut.Description = "Launch Unsloth Studio"
if ($hasValidIcon) {
$shortcut.IconLocation = "$iconPath,0"
}
$shortcut.Save()
$createdShortcutCount++
} catch {
substep "could not create shortcut at ${linkPath}: $($_.Exception.Message)" "Yellow"
}
}
if ($createdShortcutCount -gt 0) {
substep "Created Unsloth Studio shortcut"
} else {
substep "no Unsloth Studio shortcuts were created" "Yellow"
}
} catch {
substep "shortcut creation unavailable: $($_.Exception.Message)" "Yellow"
}
} catch {
substep "shortcut setup failed; skipping shortcuts: $($_.Exception.Message)" "Yellow"
}
}
# Regen .lnk + launcher only; used by `unsloth studio update`.
if ($ShortcutsOnly) {
if ($TauriMode) { return }
$UnslothExe = Join-Path $VenvDir "Scripts\unsloth.exe"
if (-not (Test-Path -LiteralPath $UnslothExe)) {
Write-Host "[ERROR] unsloth.exe missing at $UnslothExe; run install.ps1 first." -ForegroundColor Red
# throw (not Exit-InstallFailure) so non-Tauri callers see rc != 0.
throw "unsloth.exe missing"
}
New-StudioShortcuts -UnslothExePath $UnslothExe
return
}
# ── Check winget ──
# winget is only needed to install Python or uv. If both are
# already on PATH (Windows ARM64 GitHub-hosted runners, manual
# python.org + Astral uv installs, corporate locked-down hosts
# without the Store, etc.) the script can proceed without it.
# We defer the hard failure to the Python / uv install branches
# below, where winget is actually invoked.
Write-TauriLog "STEP" "Checking system dependencies"
$script:WingetAvailable = [bool](Get-Command winget -ErrorAction SilentlyContinue)
if ($script:WingetAvailable) {
step "winget" "available"
} else {
step "winget" "not available -- will require Python + uv to be already installed" "Yellow"
substep "Get it from https://aka.ms/getwinget if Python / uv are not already on PATH." "Yellow"
}
# ── Helper: detect a working Python 3.11-3.13 on the system ──
# Returns the version string (e.g. "3.13") or "" if none found.
# Uses try-catch + stderr redirection so that App Execution Alias stubs
# (WindowsApps) and other non-functional executables are probed safely
# without triggering $ErrorActionPreference = "Stop".
#
# Skips Anaconda/Miniconda Python: conda-bundled CPython ships modified
# DLL search paths that break torch's c10.dll loading on Windows.
# Standalone CPython (python.org, winget, uv) does not have this issue.
#
# NOTE: A venv created from conda Python inherits conda's base_prefix
# even if the venv path does not contain "conda". We check both the
# executable path AND sys.base_prefix to catch this.
$script:CondaSkipPattern = '(?i)(conda|miniconda|anaconda|miniforge|mambaforge)'
function Test-IsCondaPython {
param([string]$Exe)
if ($Exe -match $script:CondaSkipPattern) { return $true }
try {
$basePrefix = (& $Exe -c "import sys; print(sys.base_prefix)" 2>$null | Out-String).Trim()
if ($basePrefix -match $script:CondaSkipPattern) { return $true }
} catch { }
return $false
}
# Returns @{ Version = "3.13"; Path = "C:\...\python.exe" } or $null.
# The resolved Path is passed to `uv venv --python` to prevent uv from
# re-resolving the version string back to a conda interpreter.
function Find-CompatiblePython {
# Try the Python Launcher first (most reliable on Windows)
# py.exe resolves to the standard CPython install, not conda.
$pyLauncher = Get-Command py -CommandType Application -ErrorAction SilentlyContinue
if ($pyLauncher -and $pyLauncher.Source -notmatch $script:CondaSkipPattern) {
foreach ($minor in @("3.13", "3.12", "3.11")) {
try {
$out = & $pyLauncher.Source "-$minor" --version 2>&1 | Out-String
if ($out -match "Python (3\.1[1-3])\.\d+") {
$ver = $Matches[1]
# Resolve the actual executable path and verify it is not conda-based
$resolvedExe = (& $pyLauncher.Source "-$minor" -c "import sys; print(sys.executable)" 2>$null | Out-String).Trim()
if ($resolvedExe -and (Test-Path $resolvedExe) -and -not (Test-IsCondaPython $resolvedExe)) {
return @{ Version = $ver; Path = $resolvedExe }
}
}
} catch {}
}
}
# Try python3 / python via Get-Command -All to look past stubs that
# might shadow a real Python further down PATH.
# Skip WindowsApps entries: the App Execution Alias stubs live there
# and can open the Microsoft Store as a side effect. Legitimate Store
# Python is already detected via the py launcher above (Store packages
# include py since Python 3.11).
# Skip Anaconda/Miniconda: check both path and sys.base_prefix.
foreach ($name in @("python3", "python")) {
foreach ($cmd in @(Get-Command $name -All -ErrorAction SilentlyContinue)) {
if (-not $cmd.Source) { continue }
if ($cmd.Source -like "*\WindowsApps\*") { continue }
if (Test-IsCondaPython $cmd.Source) { continue }
try {
$out = & $cmd.Source --version 2>&1 | Out-String
if ($out -match "Python (3\.1[1-3])\.\d+") {
return @{ Version = $Matches[1]; Path = $cmd.Source }
}
} catch {}
}
}
return $null
}
# ── Install Python if no compatible version (3.11-3.13) found ──
# Find-CompatiblePython returns @{ Version = "3.13"; Path = "C:\...\python.exe" } or $null.
Write-TauriLog "STEP" "Installing Python"
$DetectedPython = Find-CompatiblePython
if ($DetectedPython) {
step "python" "Python $($DetectedPython.Version) already installed"
}
if (-not $DetectedPython) {
if (-not $script:WingetAvailable) {
Write-Host "[ERROR] No compatible Python (3.11-3.13) found and winget is unavailable on this host." -ForegroundColor Red
Write-Host " Install Python $PythonVersion from https://www.python.org/downloads/" -ForegroundColor Yellow
Write-Host " and re-run this installer (make sure 'Add Python to PATH' is checked)." -ForegroundColor Yellow
return (Exit-InstallFailure "winget required to install Python on this host")
}
substep "installing Python ${PythonVersion}..."
$pythonPackageId = "Python.Python.$PythonVersion"
# Temporarily lower ErrorActionPreference so that winget stderr
# (progress bars, warnings) does not become a terminating error
# on PowerShell 5.1 where native-command stderr is ErrorRecord.
$prevEAP = $ErrorActionPreference
$ErrorActionPreference = "Continue"
try {
winget install -e --id $pythonPackageId --accept-package-agreements --accept-source-agreements
$wingetExit = $LASTEXITCODE
} catch { $wingetExit = 1 }
$ErrorActionPreference = $prevEAP
Refresh-SessionPath
# Re-detect after install (PATH may have changed)
$DetectedPython = Find-CompatiblePython
if (-not $DetectedPython) {
# Python still not functional after winget -- force reinstall.
# This handles both real failures AND "already installed" codes where
# winget thinks Python is present but it's not actually on PATH
# (e.g. user partially uninstalled, or installed via a different method).
substep "Python not found on PATH after winget. Retrying with --force..." "Yellow"
$ErrorActionPreference = "Continue"
try {
winget install -e --id $pythonPackageId --accept-package-agreements --accept-source-agreements --force
$wingetExit = $LASTEXITCODE
} catch { $wingetExit = 1 }
$ErrorActionPreference = $prevEAP
Refresh-SessionPath
$DetectedPython = Find-CompatiblePython
}
if (-not $DetectedPython) {
Write-Host "[ERROR] Python installation failed (exit code $wingetExit)" -ForegroundColor Red
Write-Host " Please install Python $PythonVersion manually from https://www.python.org/downloads/" -ForegroundColor Yellow
Write-Host " Make sure to check 'Add Python to PATH' during installation." -ForegroundColor Yellow
Write-Host " Then re-run this installer." -ForegroundColor Yellow
return (Exit-InstallFailure "Python installation failed")
}
}
$DiagPythonVersion = $PythonVersion
if ($DetectedPython) { $DiagPythonVersion = $DetectedPython.Version }
$InitialGpuBranch = "unknown"
if ($SkipTorch) { $InitialGpuBranch = "no_torch" }
Write-TauriDiag -GpuBranch $InitialGpuBranch -TorchIndexFamily "none" -PythonVersionForDiag $DiagPythonVersion
# ── Install uv if not present ──
Write-TauriLog "STEP" "Installing uv package manager"
if (-not (Get-Command uv -ErrorAction SilentlyContinue)) {
substep "installing uv package manager..."
if ($script:WingetAvailable) {
$prevEAP = $ErrorActionPreference
$ErrorActionPreference = "Continue"
try { winget install --id=astral-sh.uv -e --accept-package-agreements --accept-source-agreements } catch {}
$ErrorActionPreference = $prevEAP
Refresh-SessionPath
}
# Fallback: if winget is unavailable or didn't put uv on PATH,
# use Astral's official PowerShell installer. This is the only
# supported path on hosts without winget (Windows ARM64 runners,
# corporate machines without the Store, etc.).
if (-not (Get-Command uv -ErrorAction SilentlyContinue)) {
substep "installing uv via https://astral.sh/uv/install.ps1..." "Yellow"
Invoke-Expression (Invoke-RestMethod -Uri "https://astral.sh/uv/install.ps1")
Refresh-SessionPath
}
}
if (-not (Get-Command uv -ErrorAction SilentlyContinue)) {
step "uv" "could not be installed" "Red"
substep "Install it from https://docs.astral.sh/uv/" "Yellow"
return (Exit-InstallFailure "uv could not be installed")
}
# ── Create venv (migrate old layout if possible, otherwise fresh) ──
# Pass the resolved executable path to uv so it does not re-resolve
# a version string back to a conda interpreter.
Write-TauriLog "STEP" "Creating virtual environment"
if (-not (Test-Path -LiteralPath $StudioHome)) {
# .NET API: New-Item -Path treats brackets as wildcards.
[System.IO.Directory]::CreateDirectory($StudioHome) | Out-Null
}
$VenvPython = Join-Path $VenvDir "Scripts\python.exe"
$_Migrated = $false
$script:StudioVenvRollbackDir = $null
$script:StudioVenvRollbackTarget = $VenvDir
$script:StudioVenvRollbackActive = $false
function Start-StudioVenvRollback {
param([Parameter(Mandatory = $true)][string]$ExistingDir)
$stamp = Get-Date -Format "yyyyMMddHHmmss"
$candidate = Join-Path $StudioHome "unsloth_studio.rollback.$stamp.$PID"
$suffix = 0
# -LiteralPath: a custom $StudioHome may contain [ ] * ? which
# plain Test-Path / Move-Item would interpret as wildcards.
while (Test-Path -LiteralPath $candidate) {
$suffix++
$candidate = Join-Path $StudioHome "unsloth_studio.rollback.$stamp.$PID.$suffix"
}
Move-Item -LiteralPath $ExistingDir -Destination $candidate -ErrorAction Stop
$script:StudioVenvRollbackDir = $candidate
$script:StudioVenvRollbackTarget = $ExistingDir
$script:StudioVenvRollbackActive = $true
substep "previous environment preserved for rollback"
}
function Restore-StudioVenvRollback {
if (-not $script:StudioVenvRollbackActive) { return }
$backup = $script:StudioVenvRollbackDir
$target = $script:StudioVenvRollbackTarget
if (-not $backup -or -not (Test-Path -LiteralPath $backup)) {
$script:StudioVenvRollbackActive = $false
return
}
substep "restoring previous environment after failed install..." "Yellow"
try {
if (Test-Path -LiteralPath $target) {
Remove-Item -LiteralPath $target -Recurse -Force -ErrorAction SilentlyContinue
}
Move-Item -LiteralPath $backup -Destination $target -Force -ErrorAction Stop
substep "restored previous environment"
$script:StudioVenvRollbackActive = $false
$script:StudioVenvRollbackDir = $null
} catch {
Write-Host "[WARN] Could not restore previous environment from $backup to $target" -ForegroundColor Yellow
Write-Host " $($_.Exception.Message)" -ForegroundColor Yellow
}
}
function Complete-StudioVenvRollback {
if (-not $script:StudioVenvRollbackActive) { return }
$backup = $script:StudioVenvRollbackDir
if ($backup -and (Test-Path -LiteralPath $backup)) {
Remove-Item -LiteralPath $backup -Recurse -Force -ErrorAction SilentlyContinue
}
$script:StudioVenvRollbackActive = $false
$script:StudioVenvRollbackDir = $null
}
if (Test-Path -LiteralPath $VenvPython) {
# why: matching guard to the .venv branch below -- in env-mode
# $StudioHome is a user-chosen workspace, so refuse to nuke an
# existing $StudioHome\unsloth_studio that lacks Studio sentinels.
# -PathType Leaf rejects a directory at the sentinel path. Accept the
# in-VENV ownership marker so partial-install retries are not blocked.
if (
$StudioRedirectMode -eq 'env' -and
-not (Test-Path -LiteralPath (Join-Path $VenvDir ".unsloth-studio-owned") -PathType Leaf) -and
-not (Test-Path -LiteralPath (Join-Path $StudioHome "share\studio.conf") -PathType Leaf) -and
-not (Test-Path -LiteralPath (Join-Path $StudioHome "bin\unsloth.exe") -PathType Leaf)
) {
Write-Host "[ERROR] $VenvDir already exists but does not look like an Unsloth Studio install." -ForegroundColor Red
Write-Host " Move it aside or choose an empty UNSLOTH_STUDIO_HOME." -ForegroundColor Yellow
throw "Refusing to delete non-Studio venv at $VenvDir"
}
# New layout already exists -- replace only after preserving rollback copy.
substep "preserving existing environment for rollback..."
try {
Start-StudioVenvRollback -ExistingDir $VenvDir
} catch {
Write-Host "[ERROR] Could not prepare existing environment for reinstall: $($_.Exception.Message)" -ForegroundColor Red
return (Exit-InstallFailure "Could not prepare existing environment for reinstall")
}
} elseif (
$StudioRedirectMode -ne 'env' `
-and (Test-Path -LiteralPath (Join-Path $StudioHome ".venv\Scripts\python.exe"))
) {
# Old layout (~/.unsloth/studio/.venv) exists -- validate before migrating.
# Skip in env-mode so we don't blow away an unrelated .venv at the
# workspace root (e.g. user's existing project Python venv).
$OldVenv = Join-Path $StudioHome ".venv"
$OldPy = Join-Path $OldVenv "Scripts\python.exe"
substep "found legacy Studio environment, validating..."
$prevEAP2 = $ErrorActionPreference
$ErrorActionPreference = "Continue"
try {
if ($SkipTorch) {
& $OldPy -c "import sys; print(sys.executable)" 2>$null | Out-Null
} else {
& $OldPy -c "import torch; A = torch.ones((2,2)); B = A + A" 2>$null | Out-Null
}
$legacyOk = ($LASTEXITCODE -eq 0)
} catch { $legacyOk = $false }
$ErrorActionPreference = $prevEAP2
if ($legacyOk) {
substep "legacy environment is healthy -- migrating..."
Move-Item -LiteralPath $OldVenv -Destination $VenvDir -Force
substep "moved .venv -> unsloth_studio"
$_Migrated = $true
} else {
substep "legacy environment failed validation -- creating fresh environment" "Yellow"
$invalidVenv = Join-Path $StudioHome (".venv.invalid.{0}.{1}" -f (Get-Date -Format "yyyyMMddHHmmss"), $PID)
Move-Item -LiteralPath $OldVenv -Destination $invalidVenv -Force -ErrorAction SilentlyContinue
}
} elseif (
$StudioRedirectMode -ne 'env' `
-and (Test-Path -LiteralPath (Join-Path $env:USERPROFILE "unsloth_studio\Scripts\python.exe"))
) {
# CWD-relative venv from old install.ps1 -> migrate to absolute path.
# Skip in env-mode so we don't relocate the default-install venv into
# the workspace root.
$CwdVenv = Join-Path $env:USERPROFILE "unsloth_studio"
substep "found CWD-relative Studio environment, migrating to $VenvDir..."
Move-Item -LiteralPath $CwdVenv -Destination $VenvDir -Force
substep "moved ~/unsloth_studio -> ~/.unsloth/studio/unsloth_studio"
$_Migrated = $true
}
if (-not (Test-Path -LiteralPath $VenvPython)) {
step "venv" "creating Python $($DetectedPython.Version) virtual environment"
substep "$VenvDir"
$venvExit = Invoke-InstallCommand { uv venv $VenvDir --python "$($DetectedPython.Path)" }
if ($venvExit -ne 0) {
Write-Host "[ERROR] Failed to create virtual environment (exit code $venvExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to create virtual environment (exit code $venvExit)" $venvExit)
}
} else {
step "venv" "using migrated environment"
substep "$VenvDir"
}
# Mark the freshly-created venv as Studio-owned so a partial install can be
# repaired by re-running install.ps1; the env-mode deletion guard above
# accepts this marker as the primary sentinel.
if (Test-Path -LiteralPath $VenvDir -PathType Container) {
try { [System.IO.File]::WriteAllText((Join-Path $VenvDir ".unsloth-studio-owned"), "") } catch {}
}
# ── Detect GPU (robust: PATH + hardcoded fallback paths, mirrors setup.ps1) ──
$HasNvidiaSmi = $false
$NvidiaSmiExe = $null
try {
$nvSmiCmd = Get-Command nvidia-smi -ErrorAction SilentlyContinue
if ($nvSmiCmd) {
& $nvSmiCmd.Source *> $null
if ($LASTEXITCODE -eq 0) { $HasNvidiaSmi = $true; $NvidiaSmiExe = $nvSmiCmd.Source }
}
} catch {}
if (-not $HasNvidiaSmi) {
foreach ($p in @(
"$env:ProgramFiles\NVIDIA Corporation\NVSMI\nvidia-smi.exe",
"$env:SystemRoot\System32\nvidia-smi.exe"
)) {
if (Test-Path $p) {
try {
& $p *> $null
if ($LASTEXITCODE -eq 0) { $HasNvidiaSmi = $true; $NvidiaSmiExe = $p; break }
} catch {}
}
}
}
# ── AMD ROCm detection (Windows) — mirrors setup.ps1 ──
$HasROCm = $false
$HipSdkInstalled = $false # HIP SDK binary found (independent of device accessibility)
$ROCmGpuLabel = $null
$ROCmVersion = $null
$ROCmGfxArch = $null
if (-not $HasNvidiaSmi) {
# hipinfo: PATH first, then HIP_PATH/ROCM_PATH bin fallback (mirrors NVIDIA smi path resolution).
# AMD HIP SDK sets HIP_PATH but may not add the bin dir to PATH depending on install type.
$hipinfoExe = Get-Command hipinfo -ErrorAction SilentlyContinue
if (-not $hipinfoExe) {
$hipRoot = if ($env:HIP_PATH) { $env:HIP_PATH } elseif ($env:ROCM_PATH) { $env:ROCM_PATH } else { $null }
$hipEnvLabel = if ($env:HIP_PATH) { "HIP_PATH" } else { "ROCM_PATH" }
if ($hipRoot) {
$hipinfoCandidate = Join-Path $hipRoot "bin\hipinfo.exe"
if (Test-Path $hipinfoCandidate) {
Write-Host " [WARN] hipinfo not on PATH -- located via ${hipEnvLabel}: $hipinfoCandidate" -ForegroundColor Yellow
Write-Host " Add '$(Join-Path $hipRoot 'bin')' to your PATH to suppress this warning" -ForegroundColor Yellow
Write-Host " Quick fix: [Environment]::SetEnvironmentVariable('PATH',`$env:PATH+';$(Join-Path $hipRoot 'bin')','User')" -ForegroundColor Yellow
$hipinfoExe = [PSCustomObject]@{ Source = $hipinfoCandidate }
} else {
Write-Host " [WARN] ${hipEnvLabel}=$hipRoot is set but hipinfo.exe not found at $hipinfoCandidate" -ForegroundColor Yellow
Write-Host " HIP SDK install may be incomplete -- re-install from:" -ForegroundColor Yellow
Write-Host " https://rocm.docs.amd.com/en/latest/deploy/windows/index.html" -ForegroundColor Yellow
}
}
}
if ($hipinfoExe) {
$HipSdkInstalled = $true # binary found → SDK is installed regardless of device state
try {
$hipOut = & $hipinfoExe.Source 2>&1 | Out-String
if ($LASTEXITCODE -eq 0 -and $hipOut -match "(?i)gcnArchName") {
$HasROCm = $true
$_hipAllArches = @([regex]::Matches($hipOut, "(?im)^\s*gcnArchName\s*:\s*(\S+)") | ForEach-Object { ($_.Groups[1].Value -split ':')[0].Trim().ToLower() })
$_hipVisIdx = if ($env:HIP_VISIBLE_DEVICES -match '^\d') { [int]($env:HIP_VISIBLE_DEVICES -split ',')[0] } elseif ($env:ROCR_VISIBLE_DEVICES -match '^\d') { [int]($env:ROCR_VISIBLE_DEVICES -split ',')[0] } else { 0 }
if ($_hipAllArches.Count -gt 0) {
$ROCmGfxArch = if ($_hipVisIdx -lt $_hipAllArches.Count) { $_hipAllArches[$_hipVisIdx] } else { $_hipAllArches[0] }
$ROCmGpuLabel = "AMD ROCm ($ROCmGfxArch)"
} else {
$ROCmGpuLabel = "AMD ROCm"
}
} elseif ($LASTEXITCODE -ne 0) {
# hipinfo ran but returned a HIP runtime error (e.g. "no ROCm-capable device detected")
$firstLine = ($hipOut -split '\r?\n' | Where-Object { $_.Trim() } | Select-Object -First 1)
Write-Host " [WARN] hipinfo returned a HIP runtime error (exit $LASTEXITCODE)" -ForegroundColor Yellow
Write-Host " $firstLine" -ForegroundColor Yellow
Write-Host " Ensure ROCm drivers are installed: https://rocm.docs.amd.com/en/latest/deploy/windows/index.html" -ForegroundColor Yellow
}
} catch {}
}
if (-not $HasROCm) {
$amdSmiExe = Get-Command "amd-smi" -ErrorAction SilentlyContinue
if ($amdSmiExe) {
try {
$smiOut = & $amdSmiExe.Source list 2>&1 | Out-String
if ($LASTEXITCODE -eq 0 -and $smiOut -match "(?im)^GPU\s*[:\[]\s*\d") {
$HasROCm = $true
# Mirror the hipinfo path: collect all gfx tokens in enumeration
# order and pick the runtime-visible one via HIP_VISIBLE_DEVICES.
$_smiVisIdx = if ($env:HIP_VISIBLE_DEVICES -match '^\d') { [int]($env:HIP_VISIBLE_DEVICES -split ',')[0] } elseif ($env:ROCR_VISIBLE_DEVICES -match '^\d') { [int]($env:ROCR_VISIBLE_DEVICES -split ',')[0] } else { 0 }
# Attempt 1: newer amd-smi versions embed the gfx arch in list output.
$_smiGfxTokens = @([regex]::Matches($smiOut, "(?i)\b(gfx\d+[a-z]?)\b") | ForEach-Object { $_.Groups[1].Value.ToLower() })
if ($_smiGfxTokens.Count -gt 0) {
$ROCmGfxArch = if ($_smiVisIdx -lt $_smiGfxTokens.Count) { $_smiGfxTokens[$_smiVisIdx] } else { $_smiGfxTokens[0] }
$ROCmGpuLabel = "AMD ROCm ($ROCmGfxArch)"
} else {
# Attempt 2: 'static --asic' exposes ASIC details on ROCm 6+,
# including the GFX target needed for wheel index selection.
$smiAsicOut = ""
try { $smiAsicOut = & $amdSmiExe.Source static --asic 2>&1 | Out-String } catch {}
$_asicGfxTokens = @([regex]::Matches($smiAsicOut, "(?i)\b(gfx\d+[a-z]?)\b") | ForEach-Object { $_.Groups[1].Value.ToLower() })
if ($_asicGfxTokens.Count -gt 0) {
$ROCmGfxArch = if ($_smiVisIdx -lt $_asicGfxTokens.Count) { $_asicGfxTokens[$_smiVisIdx] } else { $_asicGfxTokens[0] }
$ROCmGpuLabel = "AMD ROCm ($ROCmGfxArch)"
} elseif ($smiAsicOut -match "(?im)Market.?Name\s*[:\|]\s*([^\r\n]+)") {
$ROCmGpuLabel = "AMD ROCm ($($Matches[1].Trim()))"
} else {
$ROCmGpuLabel = "AMD ROCm"
}
}
}
} catch {}
}
}
if (-not $HasROCm) {
try {
$wmiGpu = Get-WmiObject Win32_VideoController -ErrorAction SilentlyContinue |
Where-Object { $_.Name -match "AMD|Radeon" } |
Select-Object -First 1
if ($wmiGpu) { $ROCmGpuLabel = $wmiGpu.Name }
} catch {}
}
# ── Arch resolution: env-var override → name inference ──────────────
# Covers users whose amd-smi is too old to report the GFX target and
# who don't have hipinfo (HIP-runtime-only, common on Strix Halo / iGPU).
if ($HasROCm -and -not $ROCmGfxArch) {
# 1. Manual override: set UNSLOTH_ROCM_GFX_ARCH=gfx1151 before running.
if ($env:UNSLOTH_ROCM_GFX_ARCH) {
$ROCmGfxArch = $env:UNSLOTH_ROCM_GFX_ARCH.Trim().ToLower()
$ROCmGpuLabel = "AMD ROCm ($ROCmGfxArch)"
substep "gfx arch from UNSLOTH_ROCM_GFX_ARCH env override: $ROCmGfxArch" "Cyan"
}
# 2. Best-effort name → arch lookup from marketing name (amd-smi / WMI).
elseif ($ROCmGpuLabel) {
$nameArchTable = @(
@{ P = "9070 XT|9080"; A = "gfx1201" } # RDNA 4
@{ P = "9070|9060"; A = "gfx1200" } # RDNA 4
@{ P = "8060S|890M|Strix Halo|HX 37[05]|HX 38[05]|AI 9 HX"; A = "gfx1151" } # RDNA 3.5 iGPU (Strix Halo / Radeon 8060S retail)
@{ P = "880M|Strix Point|AI 9 36[05]|AI 7 35[05]|AI 5 34[05]"; A = "gfx1150" } # RDNA 3.5 iGPU (Strix Point)
@{ P = "RX 7900|RX 7800|RX 7700(?! S)"; A = "gfx1100" } # RDNA 3 desktop
@{ P = "RX 7600"; A = "gfx1102" } # RDNA 3
@{ P = "780M|760M|740M|Phoenix"; A = "gfx1103" } # RDNA 3 iGPU (Phoenix)
)
foreach ($row in $nameArchTable) {
if ($ROCmGpuLabel -match $row.P) {
$ROCmGfxArch = $row.A
$ROCmGpuLabel = "AMD ROCm ($ROCmGfxArch)"
substep "gfx arch inferred from GPU name: $ROCmGfxArch" "Cyan"
substep "Tip: set UNSLOTH_ROCM_GFX_ARCH=$ROCmGfxArch to skip inference next time" "Cyan"
break
}
}
}
}
# Capture ROCm version for wheel selection (hipconfig, then amd-smi).
# Run whenever the HIP SDK binary is present, not just when the device is accessible --
# hipconfig --version works even when hipinfo reports no ROCm device (driver issue).
if ($HasROCm -or $HipSdkInstalled) {
$hipConfigExe = Get-Command hipconfig -ErrorAction SilentlyContinue
if (-not $hipConfigExe) {
$hipRoot = if ($env:HIP_PATH) { $env:HIP_PATH } elseif ($env:ROCM_PATH) { $env:ROCM_PATH } else { $null }
if ($hipRoot) {
$hipConfigCandidate = Join-Path $hipRoot "bin\hipconfig.exe"
if (Test-Path $hipConfigCandidate) {
$hipConfigEnvLabel = if ($env:HIP_PATH) { "HIP_PATH" } else { "ROCM_PATH" }
Write-Host " [WARN] hipconfig not on PATH -- located via ${hipConfigEnvLabel}: $hipConfigCandidate" -ForegroundColor Yellow
$hipConfigExe = [PSCustomObject]@{ Source = $hipConfigCandidate }
}
}
}
if ($hipConfigExe) {
try {
$hipVerOut = & $hipConfigExe.Source --version 2>&1 | Out-String
if ($LASTEXITCODE -eq 0) {
$hipVerLine = ($hipVerOut -split '\r?\n' | Where-Object { $_.Trim() } | Select-Object -First 1).Trim()
if ($hipVerLine -match '(\d+\.\d+)') {
$ROCmVersion = $Matches[1]
$ROCmVersionFull = $hipVerLine
}
}
} catch {}
}
if (-not $ROCmVersion) {
$amdSmiVer = Get-Command "amd-smi" -ErrorAction SilentlyContinue
if ($amdSmiVer) {
try {
$smiVerOut = & $amdSmiVer.Source version 2>&1 | Out-String
if ($LASTEXITCODE -eq 0 -and $smiVerOut -match 'ROCm version:\s*(\d+\.\d+)') {
$ROCmVersion = $Matches[1]
}
} catch {}
}
}
}
}
if ($HasNvidiaSmi) {
step "gpu" "NVIDIA GPU detected"
} elseif ($HasROCm) {
step "gpu" $ROCmGpuLabel
$hipSdkPath = if ($env:HIP_PATH) { $env:HIP_PATH } elseif ($env:ROCM_PATH) { $env:ROCM_PATH } else { "on system PATH" }
substep "HIP SDK: $hipSdkPath"
if ($ROCmVersionFull) { substep "hipconfig: $ROCmVersionFull" }
} elseif ($HipSdkInstalled -and $ROCmGpuLabel) {
# HIP SDK is installed but ROCm can't see the device (driver issue, not SDK issue)
$sdkVer = if ($ROCmVersionFull) { " (HIP $ROCmVersionFull)" } else { "" }
step "gpu" "AMD GPU detected -- not ROCm-accessible$sdkVer" "Yellow"
substep "Detected: $ROCmGpuLabel" "Yellow"
substep "[WARN] HIP SDK is installed but hipinfo reports no ROCm-capable device." "Yellow"
substep " This is a driver issue, not an SDK issue." "Yellow"
substep " Ensure the ROCm compute driver is installed alongside the display driver:" "Yellow"
substep " https://rocm.docs.amd.com/en/latest/deploy/windows/index.html" "Yellow"
} elseif ($ROCmGpuLabel) {
step "gpu" "AMD GPU detected -- HIP SDK not found" "Yellow"
substep "Detected: $ROCmGpuLabel" "Yellow"
substep "Install the HIP SDK for ROCm GPU inference:" "Yellow"
substep "https://rocm.docs.amd.com/en/latest/deploy/windows/index.html" "Yellow"
} else {
step "gpu" "none (chat-only / GGUF)" "Yellow"
substep "Training and GPU inference require an NVIDIA or AMD ROCm GPU." "Yellow"
}
# ── Choose the correct PyTorch index URL based on driver CUDA version ──
# Mirrors Get-PytorchCudaTag in setup.ps1.
function Get-TorchIndexUrl {
$baseUrl = if ($env:UNSLOTH_PYTORCH_MIRROR) { $env:UNSLOTH_PYTORCH_MIRROR.TrimEnd('/') } else { "https://download.pytorch.org/whl" }
if (-not $NvidiaSmiExe) { return "$baseUrl/cpu" }
try {
$output = & $NvidiaSmiExe 2>&1 | Out-String
# Newer NVIDIA drivers (e.g. 610.x on Windows) print
# "CUDA UMD Version: X.Y" instead of the legacy "CUDA Version: X.Y".
# Accept both spellings so we don't fall through to the cu126 default.
if ($output -match 'CUDA(?: UMD)? Version:\s+(\d+)\.(\d+)') {
$major = [int]$Matches[1]; $minor = [int]$Matches[2]
if ($major -ge 13) { return "$baseUrl/cu130" }
if ($major -eq 12 -and $minor -ge 8) { return "$baseUrl/cu128" }
if ($major -eq 12 -and $minor -ge 6) { return "$baseUrl/cu126" }
if ($major -ge 12) { return "$baseUrl/cu124" }
if ($major -ge 11) { return "$baseUrl/cu118" }
return "$baseUrl/cpu"
}
} catch {}
substep "could not determine CUDA version from nvidia-smi, defaulting to cu126" "Yellow"
return "$baseUrl/cu126"
}
$TorchIndexUrl = Get-TorchIndexUrl
# ── GPU arch → newest compatible Windows ROCm wheel release ──
# Wheels bundle their own ROCm runtime; the installed HIP SDK version does
# not constrain which release to use. Always picks the newest release that
# supports the GPU architecture.
# ── AMD Windows ROCm: arch-aware pip index (repo.amd.com) ──
# Wheels bundle their own ROCm runtime and support all Python versions.
# Override with UNSLOTH_ROCM_WINDOWS_MIRROR for air-gapped / mirror installs.
$ROCmIndexUrl = $null
$ROCmTorchFloor = $null
if ($HasROCm -and $TorchIndexUrl -like "*/cpu" -and -not $SkipTorch) {
$amdIndexBase = if ($env:UNSLOTH_ROCM_WINDOWS_MIRROR) { $env:UNSLOTH_ROCM_WINDOWS_MIRROR.TrimEnd('/') } else { "https://repo.amd.com/rocm/whl" }
$archFamilyMap = @{
"gfx1201" = "gfx120X-all"; "gfx1200" = "gfx120X-all" # RDNA 4
"gfx1151" = "gfx1151"; "gfx1150" = "gfx1150" # RDNA 3.5 (Strix Halo/Point)
"gfx1103" = "gfx110X-all"; "gfx1102" = "gfx110X-all" # RDNA 3
"gfx1101" = "gfx110X-all"; "gfx1100" = "gfx110X-all"
"gfx90a" = "gfx90a"; "gfx908" = "gfx908" # MI200/MI100
}
# gfx120X (RDNA 4) and gfx1151/gfx1150 (Strix) have a null-pointer bug in
# torch._C._grouped_mm on torch <2.11.0 (rocm7.12 and rocm7.1 respectively).
# TheRock issues #5284 and #3284. Force torch>=2.11.0 so pip never resolves
# to the broken 2.10.0 wheels even though they exist on the AMD index.
# The <2.12.0 ceiling matches the Linux install_python_stack.py constraint
# for the same arches: AMD actively publishes new versions on their index,
# so without a ceiling a future 2.12.0+rocmX.Y wheel would be pulled in
# automatically before it has been validated on these architectures.
# Bump the ceiling here (and in install_python_stack.py) when 2.12.x is
# confirmed working on gfx120X / Strix.
$torchFloorMap = @{
"gfx1201" = "torch>=2.11.0,<2.12.0"; "gfx1200" = "torch>=2.11.0,<2.12.0"
"gfx1151" = "torch>=2.11.0,<2.12.0"; "gfx1150" = "torch>=2.11.0,<2.12.0"
}
$archFamily = if ($ROCmGfxArch -and $archFamilyMap.ContainsKey($ROCmGfxArch)) { $archFamilyMap[$ROCmGfxArch] } else { $null }
if ($archFamily) {
$ROCmIndexUrl = "$amdIndexBase/$archFamily/"
$ROCmTorchFloor = if ($ROCmGfxArch -and $torchFloorMap.ContainsKey($ROCmGfxArch)) { $torchFloorMap[$ROCmGfxArch] } else { $null }
$archLabel = if ($ROCmGfxArch) { $ROCmGfxArch } else { "AMD GPU" }
substep "$archLabel -- AMD repo.amd.com index selected" "Cyan"
if ($ROCmTorchFloor) {
substep " enforcing $ROCmTorchFloor (known _grouped_mm bug in older wheels)" "Cyan"
}
} elseif ($ROCmGfxArch) {
substep "AMD GPU ($ROCmGfxArch) not in supported arch list -- falling back to CPU-only PyTorch" "Yellow"
} else {
substep "AMD GPU detected but arch unknown -- falling back to CPU-only PyTorch" "Yellow"
}
}
if ($ROCmIndexUrl) {
$TorchIndexFamily = "rocm"
} else {
$TorchIndexFamily = Get-TauriTorchIndexFamily $TorchIndexUrl
}
$GpuBranch = Get-TauriGpuBranch $TorchIndexFamily
Write-TauriDiag -GpuBranch $GpuBranch -TorchIndexFamily $TorchIndexFamily -PythonVersionForDiag $DetectedPython.Version
# ── Print CPU-only hint when no GPU detected ──
if (-not $SkipTorch -and -not $ROCmIndexUrl -and $TorchIndexUrl -like "*/cpu") {
Write-Host ""
if ($HipSdkInstalled -and -not $HasROCm) {
substep "Installing CPU-only PyTorch (HIP SDK found but GPU not ROCm-accessible)." "Yellow"
} elseif ($ROCmGpuLabel) {
substep "Installing CPU-only PyTorch (ROCm wheels require the HIP SDK)." "Yellow"
} else {
substep "No NVIDIA GPU detected." "Yellow"
}
substep "Installing CPU-only PyTorch. If you only need GGUF chat/inference," "Yellow"
substep "re-run with --no-torch for a faster, lighter install:" "Yellow"
substep ".\install.ps1 --no-torch" "Yellow"
Write-Host ""
}
# ── Install PyTorch first, then unsloth separately ──
#
# Why two steps?
# `uv pip install unsloth --torch-backend=cpu` on Windows resolves to
# unsloth==2024.8 (a pre-CLI release with no unsloth.exe) because the
# cpu-only solver cannot satisfy newer unsloth's dependencies.
# Installing torch first from the explicit CUDA index, then upgrading
# unsloth in a second step, avoids this solver dead-end.
#
# Why --upgrade-package instead of --upgrade?
# `--upgrade unsloth` re-resolves ALL dependencies including torch,
# pulling torch from default PyPI and stripping the +cuXXX suffix
# that step 1 installed (e.g. torch 2.5.1+cu124 -> 2.10.0 with no
# CUDA suffix). `--upgrade-package unsloth` upgrades ONLY unsloth
# to the latest version while preserving the already-pinned torch
# CUDA wheels. Missing dependencies (transformers, trl, peft, etc.)
# are still pulled in because they are new, not upgrades.
#
# ── Helper: find no-torch-runtime.txt ──
function Find-NoTorchRuntimeFile {
if ($StudioLocalInstall -and (Test-Path (Join-Path $RepoRoot "studio\backend\requirements\no-torch-runtime.txt"))) {
return Join-Path $RepoRoot "studio\backend\requirements\no-torch-runtime.txt"
}
$installed = Get-ChildItem -LiteralPath $VenvDir -Recurse -Filter "no-torch-runtime.txt" -ErrorAction SilentlyContinue |
Where-Object { $_.FullName -like "*studio*backend*requirements*no-torch-runtime.txt" } |
Select-Object -ExpandProperty FullName -First 1
return $installed
}
if ($_Migrated) {
# Migrated env: force-reinstall unsloth+unsloth-zoo to ensure clean state
# in the new venv location, while preserving existing torch/CUDA
Write-TauriLog "STEP" "Installing unsloth"
substep "upgrading unsloth in migrated environment..."
if ($SkipTorch) {
# No-torch: install unsloth + unsloth-zoo with --no-deps, then
# runtime deps (typer, safetensors, transformers, etc.) with --no-deps.
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps --reinstall-package unsloth --reinstall-package unsloth-zoo "unsloth>=2026.5.8" unsloth-zoo }
if ($baseInstallExit -eq 0) {
# Resolve pydantic WITH deps so pip pins pydantic-core
# to the matching version (no-torch-runtime.txt below
# is --no-deps). All transitive deps are torch-free.
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython pydantic }
}
if ($baseInstallExit -eq 0) {
$NoTorchReq = Find-NoTorchRuntimeFile
if ($NoTorchReq) {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps -r $NoTorchReq }
}
}
} else {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --reinstall-package unsloth --reinstall-package unsloth-zoo "unsloth>=2026.5.8" unsloth-zoo }
}
if ($baseInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install unsloth (exit code $baseInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install unsloth (exit code $baseInstallExit)" $baseInstallExit)
}
if ($StudioLocalInstall) {
substep "overlaying local repo (editable)..."
$overlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython -e $RepoRoot --no-deps }
if ($overlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay local repo (exit code $overlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay local repo (exit code $overlayExit)" $overlayExit)
}
substep "overlaying unsloth-zoo from git main..."
$zooOverlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps --reinstall-package unsloth-zoo "unsloth-zoo @ git+https://github.com/unslothai/unsloth-zoo" }
if ($zooOverlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" $zooOverlayExit)
}
}
} elseif ($TorchIndexUrl -or $ROCmIndexUrl) {
if ($SkipTorch) {
substep "skipping PyTorch (--no-torch flag set)." "Yellow"
} elseif ($ROCmIndexUrl) {
Write-TauriLog "STEP" "Installing PyTorch (AMD ROCm Windows)"
substep "installing PyTorch from $ROCmIndexUrl..."
$torchSpec = if ($ROCmTorchFloor) { $ROCmTorchFloor } else { "torch" }
$torchInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --force-reinstall --index-url $ROCmIndexUrl $torchSpec torchvision torchaudio }
if ($torchInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install AMD ROCm PyTorch (exit code $torchInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install AMD ROCm PyTorch (exit code $torchInstallExit)" $torchInstallExit)
}
} else {
Write-TauriLog "STEP" "Installing PyTorch"
substep "installing PyTorch ($TorchIndexUrl)..."
$torchInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython "torch>=2.4,<2.11.0" torchvision torchaudio --index-url $TorchIndexUrl }
if ($torchInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install PyTorch (exit code $torchInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install PyTorch (exit code $torchInstallExit)" $torchInstallExit)
}
}
Write-TauriLog "STEP" "Installing unsloth"
substep "installing unsloth (this may take a few minutes)..."
if ($SkipTorch) {
# No-torch: install unsloth + unsloth-zoo with --no-deps, then
# runtime deps (typer, safetensors, transformers, etc.) with --no-deps.
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps --upgrade-package unsloth --upgrade-package unsloth-zoo "unsloth>=2026.5.8" unsloth-zoo }
if ($baseInstallExit -eq 0) {
# Same pydantic-with-deps trick as the migrated branch.
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython pydantic }
}
if ($baseInstallExit -eq 0) {
$NoTorchReq = Find-NoTorchRuntimeFile
if ($NoTorchReq) {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps -r $NoTorchReq }
}
}
} elseif ($StudioLocalInstall) {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --upgrade-package unsloth "unsloth>=2026.5.8" unsloth-zoo }
} else {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --upgrade-package unsloth -- "$PackageName" }
}
if ($baseInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install unsloth (exit code $baseInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install unsloth (exit code $baseInstallExit)" $baseInstallExit)
}
if ($StudioLocalInstall) {
substep "overlaying local repo (editable)..."
$overlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython -e $RepoRoot --no-deps }
if ($overlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay local repo (exit code $overlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay local repo (exit code $overlayExit)" $overlayExit)
}
substep "overlaying unsloth-zoo from git main..."
$zooOverlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps --reinstall-package unsloth-zoo "unsloth-zoo @ git+https://github.com/unslothai/unsloth-zoo" }
if ($zooOverlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" $zooOverlayExit)
}
}
} else {
# Fallback: GPU detection failed to produce a URL -- let uv resolve torch
Write-TauriLog "STEP" "Installing unsloth"
substep "installing unsloth (this may take a few minutes)..."
if ($StudioLocalInstall) {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython unsloth-zoo "unsloth>=2026.5.8" --torch-backend=auto }
if ($baseInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install unsloth (exit code $baseInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install unsloth (exit code $baseInstallExit)" $baseInstallExit)
}
substep "overlaying local repo (editable)..."
$overlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython -e $RepoRoot --no-deps }
if ($overlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay local repo (exit code $overlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay local repo (exit code $overlayExit)" $overlayExit)
}
substep "overlaying unsloth-zoo from git main..."
$zooOverlayExit = Invoke-InstallCommand { uv pip install --python $VenvPython --no-deps --reinstall-package unsloth-zoo "unsloth-zoo @ git+https://github.com/unslothai/unsloth-zoo" }
if ($zooOverlayExit -ne 0) {
Write-Host "[ERROR] Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to overlay unsloth-zoo (exit code $zooOverlayExit)" $zooOverlayExit)
}
} else {
$baseInstallExit = Invoke-InstallCommand { uv pip install --python $VenvPython --torch-backend=auto -- "$PackageName" }
if ($baseInstallExit -ne 0) {
Write-Host "[ERROR] Failed to install unsloth (exit code $baseInstallExit)" -ForegroundColor Red
return (Exit-InstallFailure "Failed to install unsloth (exit code $baseInstallExit)" $baseInstallExit)
}
}
}
# Overlay Tauri-bundled studio fixes that may be ahead of PyPI. Skipped
# for --local: the editable install above already makes _PACKAGE_ROOT in
# unsloth_cli/commands/studio.py resolve to the repo (PEP 660 __file__).
# Source paths match the Tauri bundle layout in studio/src-tauri/tauri.conf.json,
# which bundles install_python_stack.py at the bundle root next to install.ps1.
if ($TauriMode) {
$rawPath = if ($PSCommandPath) { $PSCommandPath } else { $MyInvocation.ScriptName }
if ($rawPath) {
# Strip leading \\?\ extended-length prefix if the launcher passed one.
$scriptDir = Split-Path -Parent ($rawPath -replace '^\\\\\?\\', '')
$overlayMap = [ordered]@{
"install_python_stack.py" = "Lib\site-packages\studio\install_python_stack.py"
}
foreach ($rel in $overlayMap.Keys) {
$src = Join-Path $scriptDir $rel
$dst = Join-Path $VenvDir $overlayMap[$rel]
# -LiteralPath: $VenvDir derives from $StudioHome which may
# contain [ ] * ? when the user overrode UNSLOTH_STUDIO_HOME.
if (-not (Test-Path -LiteralPath $src)) { continue }
$dstParent = Split-Path -Parent $dst
if (-not (Test-Path -LiteralPath $dstParent)) {
Write-Host "[WARN] Overlay target dir missing: $dstParent; studio setup may use stale bundled file" -ForegroundColor Yellow
continue
}
try {
if (-not (Test-Path -LiteralPath $dst)) {
# Backfill: target file missing but parent dir exists.
Copy-Item -LiteralPath $src -Destination $dst -Force
substep ("backfilled bundled " + (Split-Path -Leaf $rel))
} else {
# Hash-compare so re-runs are no-ops when files already match.
$srcHash = (Get-FileHash -LiteralPath $src -Algorithm SHA256).Hash
$dstHash = (Get-FileHash -LiteralPath $dst -Algorithm SHA256).Hash
if ($srcHash -ne $dstHash) {
Copy-Item -LiteralPath $src -Destination $dst -Force
substep ("applied bundled " + (Split-Path -Leaf $rel))
}
}
} catch {
Write-Host "[WARN] Could not overlay $($rel): $($_.Exception.Message); studio setup may use stale bundled file" -ForegroundColor Yellow
}
}
}
}
# ── Run studio setup ──
# setup.ps1 will handle installing Git, CMake, Visual Studio Build Tools,
# CUDA Toolkit, Node.js, and other dependencies automatically via winget.
Write-TauriLog "STEP" "Running studio setup"
step "setup" "running unsloth studio setup..."
$UnslothExe = Join-Path $VenvDir "Scripts\unsloth.exe"
if (-not (Test-Path -LiteralPath $UnslothExe)) {
Write-TauriLog "ERROR" "unsloth CLI was not installed correctly"
Write-Host "[ERROR] unsloth CLI was not installed correctly." -ForegroundColor Red
Write-Host " Expected: $UnslothExe" -ForegroundColor Yellow
Write-Host " This usually means an older unsloth version was installed that does not include the Studio CLI." -ForegroundColor Yellow
Write-Host " Try re-running the installer or see: https://github.com/unslothai/unsloth?tab=readme-ov-file#-quickstart" -ForegroundColor Yellow
return (Exit-InstallFailure "unsloth CLI was not installed correctly")
}
# Tell setup.ps1 to skip base package installation (install.ps1 already did it)
$env:SKIP_STUDIO_BASE = "1"
$env:STUDIO_PACKAGE_NAME = $PackageName
$env:UNSLOTH_NO_TORCH = if ($SkipTorch) { "true" } else { "false" }
# Tauri desktop app bundles its own frontend — skip Node/npm/frontend build
$env:SKIP_STUDIO_FRONTEND = if ($TauriMode) { "1" } else { "0" }
# Always set STUDIO_LOCAL_INSTALL explicitly to avoid stale values from
# a previous --local run in the same PowerShell session.
if ($StudioLocalInstall) {
$env:STUDIO_LOCAL_INSTALL = "1"
$env:STUDIO_LOCAL_REPO = $RepoRoot
} else {
$env:STUDIO_LOCAL_INSTALL = "0"
Remove-Item Env:STUDIO_LOCAL_REPO -ErrorAction SilentlyContinue
}
# Use 'studio setup' (not 'studio update') because 'update' pops
# SKIP_STUDIO_BASE, which would cause redundant package reinstallation
# and bypass the fast-path version check from PR #4667.
# Propagate UNSLOTH_STUDIO_HOME only for env-override installs; otherwise
# an inherited value would put llama.cpp in the wrong place.
$previousUnslothStudioHome = $env:UNSLOTH_STUDIO_HOME
$hadPreviousUnslothStudioHome = ($null -ne $previousUnslothStudioHome)
if ($StudioRedirectMode -eq 'env') {
$env:UNSLOTH_STUDIO_HOME = $StudioHome
} else {
Remove-Item Env:UNSLOTH_STUDIO_HOME -ErrorAction SilentlyContinue
}
$studioArgs = @('studio', 'setup')
if ($script:UnslothVerbose) { $studioArgs += '--verbose' }
$env:UNSLOTH_INSTALL_ROLLBACK_MANAGED = "1"
try {
& $UnslothExe @studioArgs
$setupExit = $LASTEXITCODE
} finally {
if ($hadPreviousUnslothStudioHome) {
$env:UNSLOTH_STUDIO_HOME = $previousUnslothStudioHome
} else {
Remove-Item Env:UNSLOTH_STUDIO_HOME -ErrorAction SilentlyContinue
}
Remove-Item Env:UNSLOTH_INSTALL_ROLLBACK_MANAGED -ErrorAction SilentlyContinue
}
if ($setupExit -ne 0) {
Write-Host "[ERROR] unsloth studio setup failed (exit code $setupExit)" -ForegroundColor Red
return (Exit-InstallFailure "unsloth studio setup failed (exit code $setupExit)" $setupExit)
}
# ── Expose `unsloth` via a shim dir containing only unsloth.exe ──
# We do NOT add the venv Scripts dir to PATH (it also holds python.exe
# and pip.exe, which would hijack the user's system interpreter).
# Hardlink preferred; falls back to copy if cross-volume or non-NTFS.
#
# Remove the legacy venv Scripts PATH entry that older installers wrote.
$LegacyScriptsDir = Join-Path $VenvDir "Scripts"
try {
$legacyKey = [Microsoft.Win32.Registry]::CurrentUser.CreateSubKey('Environment')
try {
$rawPath = $legacyKey.GetValue('Path', '', [Microsoft.Win32.RegistryValueOptions]::DoNotExpandEnvironmentNames)
if ($rawPath) {
[string[]]$pathEntries = $rawPath -split ';'
$normalLegacy = $LegacyScriptsDir.Trim().Trim('"').TrimEnd('\').ToLowerInvariant()
$expNormalLegacy = [Environment]::ExpandEnvironmentVariables($LegacyScriptsDir).Trim().Trim('"').TrimEnd('\').ToLowerInvariant()
$filtered = @($pathEntries | Where-Object {
$stripped = $_.Trim().Trim('"')
$rawNorm = $stripped.TrimEnd('\').ToLowerInvariant()
$expNorm = [Environment]::ExpandEnvironmentVariables($stripped).TrimEnd('\').ToLowerInvariant()
($rawNorm -ne $normalLegacy -and $rawNorm -ne $expNormalLegacy) -and
($expNorm -ne $normalLegacy -and $expNorm -ne $expNormalLegacy)
})
$cleanedPath = $filtered -join ';'
if ($cleanedPath -ne $rawPath) {
$legacyKey.SetValue('Path', $cleanedPath, [Microsoft.Win32.RegistryValueKind]::ExpandString)
try {
$d = "UnslothPathRefresh_$([guid]::NewGuid().ToString('N').Substring(0,8))"
[Environment]::SetEnvironmentVariable($d, '1', 'User')
[Environment]::SetEnvironmentVariable($d, [NullString]::Value, 'User')
} catch { }
}
}
} finally {
$legacyKey.Close()
}
} catch { }
$ShimDir = Join-Path $StudioHome "bin"
[System.IO.Directory]::CreateDirectory($ShimDir) | Out-Null
$ShimExe = Join-Path $ShimDir "unsloth.exe"
# Fatal preflight outside the lock-handling try/catch -- a directory at
# the shim path must not be downgraded to "Continuing with the existing
# launcher", or the install finishes with no usable shim.
if (Test-Path -LiteralPath $ShimExe -PathType Container) {
Write-Host "[ERROR] Cannot create unsloth launcher: $ShimExe is a directory." -ForegroundColor Red
Write-Host " Move or remove it manually, then re-run the installer." -ForegroundColor Yellow
throw "Cannot create unsloth launcher: $ShimExe is a directory."
}
# try/catch: if unsloth.exe is locked (Studio running), keep the old shim.
$shimUpdated = $false
try {
if (Test-Path -LiteralPath $ShimExe) { Remove-Item -LiteralPath $ShimExe -Force -ErrorAction Stop }
try {
# New-Item -ItemType HardLink does NOT accept -LiteralPath in any
# PowerShell version, so use -Path. Wildcards in $ShimExe (e.g.
# brackets in custom roots) glob-expand here and fall through to
# the Copy-Item -LiteralPath fallback below.
New-Item -ItemType HardLink -Path $ShimExe -Target $UnslothExe -ErrorAction Stop | Out-Null
} catch {
Copy-Item -LiteralPath $UnslothExe -Destination $ShimExe -Force -ErrorAction Stop # fallback: copy
}
$shimUpdated = $true
} catch {
if (Test-Path -LiteralPath $ShimExe) {
Write-Host "[WARN] Could not refresh unsloth launcher at $ShimExe." -ForegroundColor Yellow
Write-Host " This usually means a running 'unsloth studio' process still holds the file open." -ForegroundColor Yellow
Write-Host " Close Studio and re-run the installer to pick up the latest launcher." -ForegroundColor Yellow
Write-Host " Continuing with the existing launcher." -ForegroundColor Yellow
} else {
Write-Host "[WARN] Could not create unsloth launcher at $ShimExe" -ForegroundColor Yellow
Write-Host " $($_.Exception.Message)" -ForegroundColor Yellow
Write-Host " Launch unsloth studio directly via '$UnslothExe' until the next successful install." -ForegroundColor Yellow
}
}
# Add to PATH only when launcher exists. Env-mode: session-only export,
# no registry change (workspace path may be deleted later).
$pathAdded = $false
if (Test-Path -LiteralPath $ShimExe) {
if ($StudioRedirectMode -ne 'env') {
$pathAdded = Add-ToUserPath -Directory $ShimDir -Position 'Prepend'
}
}
if ($shimUpdated -and $pathAdded) {
step "path" "added unsloth launcher to PATH"
}
Refresh-SessionPath # sync current session with registry
Complete-StudioVenvRollback
# Env-mode session export AFTER Refresh-SessionPath; otherwise a legacy
# User PATH entry (Machine > User > current $env:Path) would win.
if ($StudioRedirectMode -eq 'env' -and (Test-Path -LiteralPath $ShimExe)) {
$env:Path = "$ShimDir;$env:Path"
step "path" "exported $ShimDir for this session (no registry PATH change in env-override mode)"
}
# ── Tauri mode: done, skip shortcuts and auto-launch ──
if ($TauriMode) {
Write-TauriLog "DONE" ""
return
}
# New-StudioShortcuts gates the .lnk shortcuts on env-mode internally.
New-StudioShortcuts -UnslothExePath $UnslothExe
# Warn if another 'unsloth' wins on PATH (different venv, system pip).
# Mirrors install.sh; absolute path is still the most reliable launch.
# Uses content-hash equality (Get-FileHash) so hardlinks, symlinks, and
# identical copies of the installer's shim don't false-trigger. CommandType
# Application restricts the probe to real executables (skips aliases,
# functions, scripts).
try {
$_pathCmd = Get-Command unsloth -CommandType Application -ErrorAction SilentlyContinue | Select-Object -First 1
if ($_pathCmd) {
$_pathExe = $_pathCmd.Source
$_installedHash = (Get-FileHash -LiteralPath $UnslothExe -Algorithm SHA256 -ErrorAction SilentlyContinue).Hash
$_pathHash = (Get-FileHash -LiteralPath $_pathExe -Algorithm SHA256 -ErrorAction SilentlyContinue).Hash
if ($_installedHash -and $_pathHash -and ($_installedHash -ne $_pathHash)) {
Write-Host ""
step "warning" "another 'unsloth' wins on PATH:" "Yellow"
substep $_pathExe
substep "this installer's binary is at:"
substep $UnslothExe
substep "to use this install, call the absolute path above,"
substep "or put its dir earlier on PATH."
Write-Host ""
}
}
} catch {
# Diagnostic only; never block install on a probe failure.
}
# In interactive terminals, ask the user before starting Studio.
# In non-interactive environments (CI, Docker) just print instructions.
$IsInteractive = [Environment]::UserInteractive -and (-not [Console]::IsInputRedirected)
if ($IsInteractive) {
Write-Host ""
$reply = Read-Host " Start Unsloth Studio now? [Y/n]"
if ([string]::IsNullOrWhiteSpace($reply) -or $reply -match '^[Yy]') {
& $UnslothExe studio -p 8888
} else {
step "launch" "to start later, run:"
substep "unsloth studio -p 8888"
substep "(add -H 0.0.0.0 to allow network / cloud access)"
Write-Host ""
}
} else {
step "launch" "manual commands:"
# Single-quote the printed paths so $-vars / backticks in custom roots
# do not reparse when the user pastes the command.
$_actLiteral = "'" + ((Join-Path $VenvDir "Scripts\Activate.ps1") -replace "'", "''") + "'"
if ($StudioRedirectMode -eq 'env') {
# Env-mode skips registry PATH; print the absolute shim path.
$_shim = Join-Path $StudioHome "bin\unsloth.exe"
$_shimLiteral = "'" + ($_shim -replace "'", "''") + "'"
substep "& $_shimLiteral studio -p 8888"
substep "or activate env first:"
substep "& $_actLiteral"
substep "unsloth studio -p 8888"
} else {
substep "& $_actLiteral"
substep "unsloth studio -p 8888"
}
substep "(add -H 0.0.0.0 to allow network / cloud access)"
Write-Host ""
}
}
Install-UnslothStudio @args