Fr4nk1in 4bd2952a37 feat: add dp attention support for Qwen 2/3 MoE models, fixes #6088 (#6121) ...

Co-authored-by: King.Zevin <zevin@mail.ustc.edu.cn>
Co-authored-by: Yi Zhang <1109276519@qq.com>

2025-05-16 14:44:10 -07:00

..

configs

model(vlm): pixtral (#5084)

2025-05-13 00:16:10 -07:00

connector

…

constrained

Support incremental streaming of logprob/token_ids between scheduler and detokenizer (#6225)

2025-05-12 14:33:38 -07:00

disaggregation

Minor improvements of TokenizerManager / health check (#6327)

2025-05-15 15:29:25 -07:00

distributed

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

entrypoints

Support precomputed multimodal features for Qwen-VL and Gemma3 models. (#6136)

2025-05-16 12:26:15 -07:00

layers

feat: add dp attention support for Qwen 2/3 MoE models, fixes #6088 (#6121)

2025-05-16 14:44:10 -07:00

lora

Fix lora bench (#6302)

2025-05-15 10:09:55 -07:00

managers

Support precomputed multimodal features for Qwen-VL and Gemma3 models. (#6136)

2025-05-16 12:26:15 -07:00

mem_cache

KV‑Cache (MHA, MLA): add missing start_layer / end_layer fields to MHATokenToKVPoolHost and MLATokenToKVPoolHost (#6016)

2025-05-09 15:50:06 -07:00

metrics

Improve structured outputs: fix race condition, server crash, metrics and style (#6188)

2025-05-11 08:36:16 -07:00

model_executor

[Feat] Support FlashMLA backend with MTP and FP8 KV cache (#6109)

2025-05-15 00:48:09 -07:00

model_loader

…

models

feat: add dp attention support for Qwen 2/3 MoE models, fixes #6088 (#6121)

2025-05-16 14:44:10 -07:00

openai_api

Handle empty input string for embedding models (#5621)

2025-05-11 23:17:15 +08:00

sampling

Skip the flaky test_stateful_custom_logit_processor (#6251)

2025-05-12 18:29:41 -07:00

speculative

[Feat] Support FlashMLA backend with MTP and FP8 KV cache (#6109)

2025-05-15 00:48:09 -07:00

_custom_ops.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

aio_rwlock.py

…

code_completion_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

conversation.py

model(vlm): pixtral (#5084)

2025-05-13 00:16:10 -07:00

custom_op.py

…

function_call_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

hf_transformers_utils.py

Minor improvements of TokenizerManager / health check (#6327)

2025-05-15 15:29:25 -07:00

mm_utils.py

…

model_parallel.py

…

patch_torch.py

…

reasoning_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

server_args.py

Minor improvements of TokenizerManager / health check (#6327)

2025-05-15 15:29:25 -07:00

torch_memory_saver_adapter.py

…

utils.py

Reduce MoE memory usage (#6147)

2025-05-15 09:38:28 -07:00

warmup.py

…