Cap torchvision/torchaudio versions, remove amdhip64.dll fallback, fix visible GPU count

- Cap torchvision<0.26.0 and torchaudio<2.11.0 alongside torch<2.11.0 in
  both install.sh and install_python_stack.py to prevent resolver from
  selecting incompatible companion packages from ROCm wheel index
- Remove amdhip64.dll fallback in Windows ROCm detection (DLL presence
  without hipinfo/amd-smi is not proof of GPU existence)
- Fix get_visible_gpu_count() to use _get_parent_visible_gpu_spec() which
  respects HIP_VISIBLE_DEVICES/ROCR_VISIBLE_DEVICES on ROCm hosts

install.sh

@@ -1110,7 +1110,7 @@ elif [ -n "$TORCH_INDEX_URL" ]; then
         substep "skipping PyTorch (--no-torch or Intel Mac x86_64)." "$C_WARN"
     else
         substep "installing PyTorch ($TORCH_INDEX_URL)..."
-        run_install_cmd "install PyTorch" uv pip install --python "$_VENV_PY" "torch>=2.4,<2.11.0" torchvision torchaudio \
+        run_install_cmd "install PyTorch" uv pip install --python "$_VENV_PY" "torch>=2.4,<2.11.0" "torchvision<0.26.0" "torchaudio<2.11.0" \
             --index-url "$TORCH_INDEX_URL"
         # AMD ROCm: install bitsandbytes with AMD support
         case "$TORCH_INDEX_URL" in

studio/backend/utils/hardware/hardware.py

@@ -1311,17 +1311,20 @@ def get_visible_gpu_count() -> int:
     if _visible_gpu_count is not None:
         return _visible_gpu_count
 
-    cuda_visible = os.environ.get("CUDA_VISIBLE_DEVICES")
-    if cuda_visible is not None:
-        # "" means zero GPUs, "0" means 1, "0,1,2" means 3
-        cuda_visible = cuda_visible.strip()
-        if cuda_visible == "" or cuda_visible == "-1":
+    # Use _get_parent_visible_gpu_spec() which already handles
+    # HIP_VISIBLE_DEVICES / ROCR_VISIBLE_DEVICES on ROCm.
+    visible_spec = _get_parent_visible_gpu_spec()
+    if visible_spec["raw"] is not None:
+        raw = visible_spec["raw"].strip()
+        if raw == "" or raw == "-1":
             _visible_gpu_count = 0
+        elif visible_spec["numeric_ids"] is not None:
+            _visible_gpu_count = len(visible_spec["numeric_ids"])
         else:
-            _visible_gpu_count = len([x for x in cuda_visible.split(",") if x.strip()])
+            _visible_gpu_count = len([x for x in raw.split(",") if x.strip()])
         return _visible_gpu_count
 
-    # CUDA_VISIBLE_DEVICES not set -- try torch, fall back to physical count
+    # No visibility env var set -- try torch, fall back to physical count
     try:
         import torch
 

studio/install_llama_prebuilt.py

@@ -1466,13 +1466,8 @@ def detect_host() -> HostInfo:
                 if _marker in _result.stdout.lower():
                     has_rocm = True
                     break
-        # Fallback: HIP runtime DLL indicates a working HIP installation
-        if not has_rocm and any(
-            Path(d).joinpath("amdhip64.dll").exists()
-            for d in os.environ.get("PATH", "").split(os.pathsep)
-            if d
-        ):
-            has_rocm = True
+        # Note: amdhip64.dll presence alone is NOT treated as GPU evidence
+        # since the HIP SDK can be installed without an AMD GPU.
 
     return HostInfo(
         system = system,

studio/install_python_stack.py

@@ -187,8 +187,8 @@ def _ensure_rocm_torch() -> None:
         "--force-reinstall",
         "--no-cache-dir",
         "torch>=2.4,<2.11.0",
-        "torchvision",
-        "torchaudio",
+        "torchvision<0.26.0",
+        "torchaudio<2.11.0",
         "--index-url",
         index_url,
         constrain = False,