xuebwang-amd d0a73620cc [ROCm][Quantization] add apply_vllm_mapper in quark config for models like gpt-oss (#28638) ...

Signed-off-by: xuebwang-amd <xuebwang@amd.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

2025-11-18 11:16:45 +08:00

..

compressed_tensors

Support DeepEP for Kimi-k2-thinking through enabling gemm selection for compressed-tensor marlin wna16 (#28574)

2025-11-12 21:40:45 -08:00

kernels

[ROCM] Fix ROCm warnings, environment flag access, and GEMM kernel naming for consistency in _aiter_ops.py (#28464)

2025-11-12 21:46:57 +00:00

quark

[ROCm][Quantization] add apply_vllm_mapper in quark config for models like gpt-oss (#28638)

2025-11-18 11:16:45 +08:00

utils

[Performance][Fix] update nvfp4 code to support renorm routing (#28569)

2025-11-16 18:02:42 -08:00

__init__.py

[quantization][config] enable override existing quant_config (#28510)

2025-11-14 01:24:10 +00:00

auto_round.py

[Feature][Quantization] auto_round support for mixed bits quantization (#23812)

2025-10-20 22:23:30 +00:00

awq_marlin.py

[Bugfix] Disable shared expert overlap if Marlin MoE is used (#28410)

2025-11-11 23:16:12 +00:00

awq_triton.py

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

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

awq.py

[Quantization] Automatically infer AWQ modules_to_not_convert field (#26909)

2025-10-21 01:49:28 +00:00

base_config.py

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

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

bitblas.py

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

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

bitsandbytes.py

[Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123)

2025-11-04 21:59:45 +08:00

deepspeedfp.py

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

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

experts_int8.py

[Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123)

2025-11-04 21:59:45 +08:00

fbgemm_fp8.py

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

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

fp8.py

[Kernels] Enable FlashInfer FP8 Blockscale on SM90 (for TEP DSR1) (#27134)

2025-11-14 08:02:44 -08:00

fp_quant.py

[Chore] Clean up pytorch helper functions in vllm.utils (#26908)

2025-10-18 09:48:22 -07:00

gguf.py

[Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123)

2025-11-04 21:59:45 +08:00

gptq_bitblas.py

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

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

gptq_marlin_24.py

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

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

gptq_marlin.py

[Bugfix] Disable shared expert overlap if Marlin MoE is used (#28410)

2025-11-11 23:16:12 +00:00

gptq.py

[Kernel] Add GPTQv2 format support for low-bit or asymmetric quantization, by adapting gptq_gemm (#26092)

2025-10-23 23:26:13 -04:00

hqq_marlin.py

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

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

inc.py

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

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

input_quant_fp8.py

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

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

ipex_quant.py

[XPU] add sym params to IPEXConfig (#28611)

2025-11-13 11:35:04 +00:00

kv_cache.py

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

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

modelopt.py

[Performance][Fix] update nvfp4 code to support renorm routing (#28569)

2025-11-16 18:02:42 -08:00

moe_wna16.py

[Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123)

2025-11-04 21:59:45 +08:00

mxfp4.py

[Bugfix] Fix GPT-OSS on AMD after #28603 (#28816)

2025-11-17 13:15:26 -05:00

petit.py

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

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

ptpc_fp8.py

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

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

qutlass_utils.py

[MISC] fix import violations for re and triton modules (#26654)

2025-10-16 03:38:27 -07:00

rtn.py

[Kernels] Isolate modular kernel code from FusedMoEMethodBase subclasses. (#27123)

2025-11-04 21:59:45 +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

Support using Int4PreshuffledTensor after loading (#26066)

2025-11-04 06:00:57 -05:00

tpu_int8.py

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

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