cmcamdy 13447279aa [XPU] Fix PD + MTP (#6495) ...

* fix pd + mtp

* fix code style

* fix PD + MTP, D get P's first token

* add anno for gpu(speculate_update)

* update draft insertv1

* fix wapper & kernel

* fix wapper

* fix code stype

2026-02-27 19:07:35 +08:00

..

__init__.py

…

dcu_model_runner.py

…

dcu_worker.py

Support setting communication groups in custom_allreduce and the all-to-all\transpose fused operator during the decoding phase. (#5917)

2026-01-12 14:09:39 +08:00

eplb.py

…

experts_manager.py

[Feature] Support redundant expert for eplb (#5918)

2026-01-09 17:13:24 +08:00

gcu_model_runner.py

[Model Runner] Prepare token count and move FA3 initialization into the graph (#6170)

2026-01-26 12:16:57 +08:00

gcu_worker.py

[Feature]Report FD statistical information (#5646)

2026-01-14 17:54:01 +08:00

gpu_model_runner.py

[XPU] Fix PD + MTP (#6495)

2026-02-27 19:07:35 +08:00

gpu_worker.py

[Graph Optimization] Support CUDAGraph for P/PD mixed Batch using SOT subgraph spliting mode (#6196)

2026-01-29 16:29:54 +08:00

hpu_model_runner.py

[Intel HPU] enable MoE EP for hpu (#5855)

2026-01-15 13:08:00 +08:00

hpu_worker.py

[Feature]Report FD statistical information (#5646)

2026-01-14 17:54:01 +08:00

iluvatar_model_runner.py

[Iluvatar][CI] Fix cannot import get_stop (#6165)

2026-02-10 16:57:23 +08:00

iluvatar_worker.py

Revert "[Optim] Remove limitation of number of kvcache blocks (#5612)" (#5702)

2025-12-23 15:41:33 +08:00

input_batch.py

[Feature] Add Deterministic Inference Support (#6476)

2026-02-26 19:31:51 -08:00

metax_model_runner.py

[Metax][Fix] fix error based pr #6493 (#6521)

2026-02-26 18:41:35 +08:00

metax_worker.py

[Metax] adapt to the latest develop (#6282)

2026-01-29 23:21:20 -08:00

model_runner_base.py

[Graph Optimization][CINN] Use CINN in PaddleOCR-VL ViT part (#5223)

2025-12-09 14:37:00 +08:00

output.py

[Optimization] Support logprob async copy (#6362)

2026-02-09 17:32:12 +08:00

tbo.py

cp 1131 tbo to develop (#6281)

2026-02-03 15:23:23 +08:00

worker_base.py

…

worker_process.py

[Optimization] Enable BF16 gate computation for GLM and Qwen (#6457)

2026-02-26 21:08:46 -08:00

xpu_model_runner.py

[XPU] Fix PD + MTP (#6495)

2026-02-27 19:07:35 +08:00

xpu_worker.py

[XPU][Graph Optimization] XPU Support CUDAGraph (#6152)

2026-01-22 14:41:56 +08:00