yangjianfengo1 93fcf7e4ec 【New Feature】W4afp8 supports per group quantization (#4272) ...

* w4afp8 支持per group

* code style

* 精度完成

* revert append attn utils

* ffn1 动态量化

* ffn2 支持动态量化

* code style

* code style

* 修改单测

* 修改单测

* fix bug

* Implement conditional parameter creation for layers

Add parameter creation for up_gate_proj_in_scale when ep_size > 1.

* code style

* fix conflict

* code style

* code style

* 修复w4aint8 精度

* fix ci

---------

Co-authored-by: yuanxiaolan <yuanxiaolan01@baidu.com>

2025-11-05 21:00:23 +08:00

..

attention

supports pd partn (#4615)

2025-11-04 16:36:35 +08:00

backends

[XPU]Support V1 loader in weight_only Model (#4808)

2025-11-05 17:09:11 +08:00

moe

【New Feature】W4afp8 supports per group quantization (#4272)

2025-11-05 21:00:23 +08:00

pool

[Feature] support pooling model dummy_run (#4345)

2025-10-17 13:30:55 +08:00

quantization

【New Feature】W4afp8 supports per group quantization (#4272)

2025-11-05 21:00:23 +08:00

sample

[Feature][Executor] GPU Model Runner Supports prompt_logprobs and max_logprobs (#4769)

2025-11-05 10:43:25 +08:00

__init__.py

[LLM] First commit the llm deployment code

2025-06-09 19:20:15 +08:00

activation.py

[Intel HPU] Support intel hpu platform (#4161)

2025-09-24 12:27:50 +08:00

embeddings.py

[Feature] support pooling model dummy_run (#4345)

2025-10-17 13:30:55 +08:00

linear.py

add qwen-2.5-7B-PRM/ernie-rm (#4319)

2025-10-20 15:31:03 +08:00

lm_head.py

[Feature] support qwen3-embedding model load (#4202)

2025-09-23 00:14:35 -07:00

mtp_linear.py

[Speculative Decoding][MTP]Support mtp in epdptp mode (#4614)

2025-10-28 16:02:47 +08:00

normalization.py

adaptive rms_norm's dtype (#3617)

2025-08-26 15:29:15 +08:00

pooler.py

[Feature] support pooling model runner (#4590)

2025-10-31 22:32:05 +08:00

rotary_embedding.py

[Feature] Support Paddle-OCR (#4396)

2025-10-24 23:34:30 +08:00

utils.py

[OPs] MoE support wfp8afp8(channelwise) and improve per_token_quant_fp8 (#4238)

2025-09-24 16:39:51 +08:00