unsloth/tests/studio/install
Daniel Han f695fbd0fe
Studio: extend the pinned Blackwell GPU fallback to CUDA 13.0 drivers (#5920)
The pinned b9360 cuda-13.1 Windows fallback for Blackwell (sm_120) only fired
when the driver advertised CUDA >= 13.1. The mainstream Blackwell branch ships
the r580 driver that reports CUDA 13.0, so those hosts missed the pin, were
gated off the in-release 13.3 build, and dropped to the CPU-only cuda-12.4
build.

b9360's binary is native sm_120a SASS (no PTX, so no JIT) and its bundled
runtime is cuda-13.1 cudart; both run on a 13.0 r580+ driver under CUDA
minor-version compatibility. Lower _PINNED_BLACKWELL_DRIVER_FLOOR to (13, 0) so
the pin covers the 13.0 branch too. The generic published-runtime gate stays
conservative (an unverified in-release 13.1 build is still gated off 13.0); only
the specific, hash-pinned, SASS-verified b9360 build is extended.

Refs #5887.
2026-06-01 06:35:28 -07:00
..
conftest.py fix/strix halo and windows AMD ROCm support (#5301) 2026-05-29 22:29:56 -07:00
smoke_test_llama_prebuilt.py Add installer test coverage for prebuilt llama.cpp changes (#4756) 2026-04-01 06:06:29 -07:00
smoke_test_parallel_studio_home.py tests/studio/install: parallel UNSLOTH_STUDIO_HOME smoke test (#5306) 2026-05-07 02:26:58 -07:00
test_install_llama_prebuilt_logic.py Studio: match the Linux llama.cpp prebuilt to the runtime cudart major (#5914) 2026-06-01 02:55:57 -07:00
test_llama_pr_force_and_source.py tests: raise pwsh/bash subprocess timeout from 10s to 60s (#5463) 2026-05-15 14:18:04 -07:00
test_macos_version_compat.py Studio: pin the last pre-macOS-26 llama.cpp prebuilt instead of walking back (#5896) 2026-05-31 06:18:30 -07:00
test_pr4562_bugfixes.py studio/setup.sh: cope with fresh CUDA toolkits like 13.3 (#5826) 2026-05-29 05:09:20 -07:00
test_rocm_support.py studio: ROCm cleanups follow-up to #5301 (#5874) 2026-05-30 03:06:47 -07:00
test_selection_logic.py Studio: extend the pinned Blackwell GPU fallback to CUDA 13.0 drivers (#5920) 2026-06-01 06:35:28 -07:00