laixin e330f2b86c [qwen3] support qwen3 ep moe (#5917) ...

Co-authored-by: sleepcoo <sleepcoo@gmail.com>

2025-04-30 09:15:21 -07:00

..

configs

[Feature] add support kimi vl model (#5383)

2025-04-29 21:31:19 -07:00

connector

feat(remote_model): support variable remote backend for model loader (#3964)

2025-03-14 00:40:44 -07:00

constrained

update llguidance to 0.7.11; adds StructTag (#4870)

2025-04-26 20:13:57 -07:00

disaggregation

[PD] support pd fake transfer for warmup (#5726)

2025-04-29 00:33:20 +08:00

distributed

Revert "Use device_id in dist init to reduce NCCL communicator warmup & creation overhead" (#5786)

2025-04-27 04:03:02 -07:00

entrypoints

Bump Flashinfer to 0.2.5 (#5870)

2025-04-29 19:50:57 -07:00

layers

Add A800 fused moe config for qwen3 235b (#5900)

2025-04-29 20:18:11 -07:00

lora

Feat: support cuda graph for LoRA (#4115)

2025-04-28 23:30:44 -07:00

managers

[Feature] add support kimi vl model (#5383)

2025-04-29 21:31:19 -07:00

mem_cache

Fuse MLA set kv cache kernel (#5748)

2025-04-26 18:42:22 -07:00

metrics

[metrics] Add in queue metrics (#4444)

2025-04-09 17:19:27 -07:00

model_executor

[fix] relax mem_fraction_static for h200 (#5893)

2025-04-29 17:01:12 -07:00

model_loader

Revert "Tiny refactor DefaultModelLoader.Source" (#5825)

2025-04-28 01:18:57 -07:00

models

[qwen3] support qwen3 ep moe (#5917)

2025-04-30 09:15:21 -07:00

openai_api

feat: support pythonic tool call and index in tool call streaming (#5725)

2025-04-29 17:30:44 -07:00

platforms

[feat] interface for platforms abstraction (#4928)

2025-03-31 00:04:21 -07:00

sampling

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

speculative

Disable flaky eagle tests (#5753)

2025-04-25 15:54:39 -07:00

_custom_ops.py

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

aio_rwlock.py

A better aio rwlock that guarantees the order (#2547)

2024-12-22 15:44:32 -08:00

code_completion_parser.py

fix: Use is not None instead of != None for None checks. (#5687)

2025-04-26 19:26:57 -07:00

conversation.py

[Feature] add support kimi vl model (#5383)

2025-04-29 21:31:19 -07:00

custom_op.py

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

function_call_parser.py

feat: support pythonic tool call and index in tool call streaming (#5725)

2025-04-29 17:30:44 -07:00

hf_transformers_utils.py

[Feature] add support kimi vl model (#5383)

2025-04-29 21:31:19 -07:00

mm_utils.py

docs: fix module docstrings and copyright headers (#2077)

2024-11-22 22:16:53 +08:00

model_parallel.py

Make torch TP composable with torchao (#2436)

2024-12-11 04:21:42 -08:00

patch_torch.py

Fix torch.compile cacheing (#5259)

2025-04-10 18:08:45 -07:00

reasoning_parser.py

Feat: add support for thinking mode via chat_template_kwargs.enable_t… (#5551)

2025-04-28 07:07:45 -07:00

server_args.py

feat: support pythonic tool call and index in tool call streaming (#5725)

2025-04-29 17:30:44 -07:00

torch_memory_saver_adapter.py

[CI] Fix ci tests (#5769)

2025-04-27 07:18:10 -07:00

utils.py

feat: update is_fa3_default_architecture (#5854)

2025-04-28 11:53:22 -07:00

warmup.py

Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)

2025-03-03 00:12:04 -08:00