ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

pip: cite the 0.50.0 ROCm work accurately in the bnb floor comment

Browse source 
The comment credited bnb PR #1887 as "the ROCm 4-bit GEMV fix" for every
AMD GPU. #1887 decouples blocksize from warp size and fixes a hardcoded
warp size of 32 in kgemm_4bit_inference_naive, which is a CDNA problem by
construction. The RDNA-side work is #1979 (fused 4-bit SIMT GEMM) and
#2012 (RDNA3/4 workgroup resonance). All three first ship in 0.50.0, so
the >=0.50.0 floor is unchanged; only the justification was wrong.
This commit is contained in:
LeoBorcherding 2026-07-28 13:48:39 -05:00
commit 910a657d80
1 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

5
pyproject.toml
View file

@ -175,8 +175,9 @@ huggingfacenotorch = [
]
amd = [
"unsloth[huggingfacenotorch]",
# bnb <= 0.49.2 NaNs at decode shape on every AMD GPU; the ROCm 4-bit GEMV
# fix (bnb PR #1887) first ships in 0.50.0, which is now on PyPI.
# 4-bit decode is unreliable on ROCm before 0.50.0, the first PyPI release
# carrying the full path: blocksize/warp decoupling (bnb #1887), fused SIMT
# GEMM on RDNA (#1979), RDNA3/4 workgroup fix (#2012).
"bitsandbytes>=0.50.0 ; ('linux' in sys_platform) and (platform_machine == 'AMD64' or platform_machine == 'x86_64' or platform_machine == 'aarch64')",
"bitsandbytes>=0.50.0 ; (sys_platform == 'win32') and (platform_machine == 'AMD64' or platform_machine == 'x86_64')",
]