COMMITS
August 19, 2026
C
Y
Accept artifact dir as argument in post_mlinter_review.py (#48106)
Yih-Dar committed
G
Improve group images by shape (#47964)
guarin committed
Y
Fix mlinter artifact path (#48088)
Yih-Dar committed
S
[Video] Fix convert_to_rgb channel slicing and alpha blending for RGBA videos (#48053)
Suryaansh Prithvijit Singh committed
August 18, 2026
Y
D
feat: add nvfp4 quantization (#47883)
drbh committed
N
Y
[Gemma3] Update integration test expected values for A10G (#48036)
Yih-Dar committed
R
Fallback from 'lanczos' to 'bicubic' when on cuda (#48026)
Raushan Turganbay committed
S
[serge] Fix 2 integration tests for model `generation` failing with `import_or_config` (other (2)) (#48061)
sergereview[bot] committed
S
[serge] Fix 2 integration tests regressed by commit b9090ae58cda (PR #47096) (#48060)
sergereview[bot] committed
D
[GPT2] Fix encoder_attention_mask being silently discarded in cross-attention (#47946)
David Quinlan committed
R
[Fix] Small FA-related test failures in CB (#47341)
Rémi Ouazan committed
H
fix: honor empty processor_kwargs={} in multimodal pipelines (#48044)
hungnnvidia committed
Y
[CircleCI] Enable CI for private forks, no-op for public repo (#48056)
Yih-Dar committed
R
:rotating_light: Leftover processors (#47924)
Raushan Turganbay committed
Q
Support `BatchFeature` in length-grouped samplers (#48034)
Quentin Gallouédec committed
J
Fix EOS for candidate generators (#47931)
Jordan Anderson committed
Y
K
fix failed test cases for muse_glimmer (#48011)
kaixuanliu committed
R
Cohere compass tests (#47895)
Raushan Turganbay committed
P
T
Moving mlinter to 0.1.4 (#47918)
Tarek Ziade committed
H
Fix sliding window cache index off-by-one on wraparound (#47708)
Hameed Ibrahim committed
August 17, 2026
S
[MoE] Fix Blackwell GPU crash with torch._grouped_mm on torch <= 2.8 (#48014)
Suryaansh Prithvijit Singh committed
S
[docs] Muse Glimmer (#47882)
Steven Liu committed
Y
R
Proper separation of tests (#47943)
Raushan Turganbay committed