LI SHENGYONG 611e223b7d [EPLB][Bugfix] EPLB support fp/bf16 (#5531) ...

### What this PR does / why we need it?
EPLB support dtype of fp/bf16.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?
w8a8_dynamic Baseline:
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 86.67 |

w8a8_dynamic eplb:
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 86.67 |

The fp16 conversation is normal.
The fp16 test is in progress.

Baseline fp16
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 86.67 |

eplb fp16
| dataset | version | metric | mode | vllm-api-general-chat |
|----- | ----- | ----- | ----- | -----|
| aime2024 | 604a78 | accuracy | gen | 83.33 |

- vLLM version: v0.13.0
- vLLM main:
45c1ca1ca1

Signed-off-by: shenchuxiaofugui <1311027364@qq.com>

2026-01-26 14:28:16 +08:00

..

fused_moe

[EPLB][Bugfix] EPLB support fp/bf16 (#5531)

2026-01-26 14:28:16 +08:00

triton

[Feature] Batch invariant torch.compile (#6107)

2026-01-26 09:15:06 +08:00

__init__.py

[Fusion] [Graph] Add qknorm rope fusion operator (#4711)

2025-12-17 08:53:44 +08:00

activation.py

[310P]: refactoring for 310p kvcache and some ops class (#6117)

2026-01-24 20:34:29 +08:00

flashcomm2_oshard_manager.py

[Feat] flashcomm2+oshard Generalized (#4723)

2026-01-10 22:57:57 +08:00

layer_shard_linear.py

[refactor] Refactor the interface for shard weight and remove the flashcomm2 o_shared interface. (#5181)

2026-01-08 09:05:02 +08:00

layernorm.py

[310P]: refactoring for 310p kvcache and some ops class (#6117)

2026-01-24 20:34:29 +08:00

linear_op.py

[Refactor] Quantization Module Refactor (#5738)

2026-01-23 14:13:47 +08:00

linear.py

[bugfix] fix Error 'ValueError: Duplicate layer name' (#5280)

2025-12-25 10:43:24 +08:00

mla.py

Drop vLLM 0.13.0 support (#6069)

2026-01-23 09:45:08 +08:00

mm_encoder_attention.py

[MM][Perf] Parallelize Q/K/V padding in AscendMMEncoderAttention for better performance (#6204)

2026-01-26 10:20:24 +08:00

register_custom_ops.py

[Ascend] perf: optimize rope embedding with triton kernel for huge performance gain (#5918)

2026-01-21 22:01:22 +08:00

rotary_embedding.py

Drop vLLM 0.13.0 support (#6069)

2026-01-23 09:45:08 +08:00

vocab_parallel_embedding.py

[Feat] Add custom Embedding tensor model parallel (#2616)

2025-12-12 14:41:20 +08:00

weight_prefetch.py

Update torch-npu version to 2.7.1 (#3896)

2025-10-31 17:16:31 +08:00