Shuai ShaoandJee Jee Li a60c66e3dc [Bugfix] Fix int32 index overflow in LoRA punica kernels at long context (#53034) ...

Signed-off-by: Shuai Shao <ss@rungalileo.io>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>

2026-08-21 10:37:06 +00:00

..

__init__.py

Fp8 lora dense kernel (#35242)

2026-03-13 19:05:08 +00:00

fp8_kernel_utils.py

Fp8 lora dense kernel (#35242)

2026-03-13 19:05:08 +00:00

fused_moe_lora_fp8_op.py

[LoRA] Make LoRA respect language_model_only (#37375)

2026-03-18 07:53:34 +00:00

fused_moe_lora_op.py

[Perf] Avoid some more unnecessary GPU<->CPU syncs (#51458)

2026-08-08 17:13:26 -07:00

kernel_utils.py

[Bugfix] Fix int32 index overflow in LoRA punica kernels at long context (#53034)

2026-08-21 10:37:06 +00:00

lora_expand_fp8_op.py

Fp8 lora dense kernel (#35242)

2026-03-13 19:05:08 +00:00

lora_expand_op.py

[LoRA] Support dual CUDA streams-Linear Layer (#35721)

2026-04-13 10:57:07 +08:00

lora_kernel_metadata.py

[Fix Bug]num_active_loras always equals to zero (#34119)

2026-03-02 23:17:46 +08:00

lora_shrink_fp8_op.py

Fp8 lora dense kernel (#35242)

2026-03-13 19:05:08 +00:00

lora_shrink_op.py

[LoRA] Support dual CUDA streams-Linear Layer (#35721)

2026-04-13 10:57:07 +08:00

README_TUNING.md

Allow markdownlint to run locally (#36398)

2026-03-08 20:05:24 -07:00

utils.py

[Perf][4/n] Eliminate various GPU<->CPU syncs (#42347)

2026-05-19 10:35:54 -04:00