Longzhi Wang 5cd17fd662 [Models] Add forward_meta to moe models' forward function (#5138) ...

* [Models] Add forward_meta to moe models' forward function

* fix missing param

* fix

* fix

* fix forward_meta

* fix test and remove chunked MoE releated in config

* fix test

* fix

* fix

2025-12-04 13:26:58 +08:00

..

attention

[PD Disaggregation] Support PD deployment of DeepSeekv3. (#5251)

2025-12-02 14:11:50 +08:00

backends

[Intel HPU] fix memory fragmentation issue due to warmup process and fix moe all_reduce issue (#5357)

2025-12-04 11:29:41 +08:00

batch_invariant_ops

[Deterministic] Move paddle version batch invariant pkg to Fastdeploy (#4763)

2025-12-01 11:25:48 +08:00

moe

[Models] Add forward_meta to moe models' forward function (#5138)

2025-12-04 13:26:58 +08:00

pool

[Feature] support reward model (#5301)

2025-12-02 14:55:31 +08:00

quantization

[BugFix] dynamic cache kv block_wise_fp8 not need create layer.cache_k_scale (#5362)

2025-12-03 05:32:59 -08:00

sample

[xpu] support mtp for xpu(mix) (#5274)

2025-12-01 11:03:14 +08:00

__init__.py

…

activation.py

…

embeddings.py

[Features] add audio request & fix embedding bug (#5201)

2025-12-01 11:12:17 +08:00

linear.py

…

lm_head.py

…

mtp_linear.py

…

normalization.py

[UNITEST] add test (#5305)

2025-12-02 17:59:01 +08:00

pooler.py

[Feature] support reward model (#5301)

2025-12-02 14:55:31 +08:00

rotary_embedding.py

…

utils.py

[Quantization] Support w4afp8 MoE dynamic quantization (#5282)

2025-12-02 18:56:16 +08:00