Jia GuoandClaude Opus 4.6 ec85340531 [Quantization] Support FP8 MoE bias for models like GPT-OSS (#34906) ...

Signed-off-by: jasperjiaguo <jasperg662@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-02-23 19:07:47 -08:00

..

compressed_tensors

[kv-cache, ct] Use compressed-tensors as a source of ground-truth for quant strategies (#34254)

2026-02-23 07:37:55 -07:00

kernels

[XPU][6/N] add xpu scaled_mm kernel (#34117)

2026-02-09 20:17:35 +08:00

quark

[Quark] Fix MoE fp8 activation scale handling on mi300 (#34386)

2026-02-19 21:27:14 -08:00

utils

[UX] More descriptive reasons in is_supported_config for MoE (#34908)

2026-02-19 15:20:52 -08:00

__init__.py

Add support for ModelOpt MXFP8 dense models (#33786)

2026-02-08 11:16:48 -08:00

awq_marlin.py

[MoE Refactor] Introduce MoERunner abstraction and move execution logic from FusedMoE to DefaultMoERunner (#32344)

2026-02-10 19:51:07 -05:00

awq_triton.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

awq.py

[Bugfix] Fix missing is_layer_skipped check for FusedMoE in AWQConfig (#32935)

2026-01-23 17:19:56 -05:00

base_config.py

[Quantization] - Added uses_meta_device_weights to quant config (#34645)

2026-02-17 23:43:44 -08:00

bitsandbytes.py

[ROCm] Enable bitsandbytes quantization support on ROCm (#34688)

2026-02-21 00:34:55 -08:00

cpu_wna16.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

experts_int8.py

[MoE Refactor] Introduce MoERunner abstraction and move execution logic from FusedMoE to DefaultMoERunner (#32344)

2026-02-10 19:51:07 -05:00

fbgemm_fp8.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

fp8.py

[Quantization] Support FP8 MoE bias for models like GPT-OSS (#34906)

2026-02-23 19:07:47 -08:00

fp_quant.py

"refactor: refactor_repeated_interfaces" (#32486)

2026-01-18 22:07:01 +08:00

gguf.py

[Refactor] Replace activation: str with MoEActivation enum (#33843)

2026-02-11 17:29:32 -08:00

gptq_marlin.py

[MoE Refactor] Introduce MoERunner abstraction and move execution logic from FusedMoE to DefaultMoERunner (#32344)

2026-02-10 19:51:07 -05:00

gptq.py

[Quantization][Deprecation] Remove BitBlas (#32683)

2026-01-28 11:06:22 +00:00

inc.py

[XPU][1/N] Deprecate ipex and switch to vllm-xpu-kernels for xpu platform (#33379)

2026-02-02 22:46:10 -08:00

input_quant_fp8.py

[Bugfix] Fix Dynamo unexpected keyword argument (#34320)

2026-02-16 01:32:30 -08:00

kv_cache.py

[Bugfix][Attention] Explicitly report support for kv_cache_dtype bfloat16 (#32795)

2026-01-22 19:05:18 +00:00

modelopt.py

Fix call to moe_mk in modelopt MoE modules (required for LoRA) (#34575)

2026-02-16 07:33:09 -08:00

moe_wna16.py

[Refactor] Replace activation: str with MoEActivation enum (#33843)

2026-02-11 17:29:32 -08:00

mxfp4.py

[Minor] Add logging when using MXFP4 MXFP8 TRTLLM backend (#34916)

2026-02-19 22:07:57 -08:00

petit.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

ptpc_fp8.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

qutlass_utils.py

"refactor: refactor_repeated_interfaces" (#32486)

2026-01-18 22:07:01 +08:00

schema.py

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)

2025-10-12 09:51:31 -07:00

torchao.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00