Elfie Guo 6fc9357503 [2/2] Add python wrapper for CUTLASS FP8 Blockscale MoE Kernel. (#5694)

2025-05-16 13:14:07 -07:00

..

configs

model(vlm): pixtral (#5084)

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

connector

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

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

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

[2/2] Add python wrapper for CUTLASS FP8 Blockscale MoE Kernel. (#5694)

2025-05-16 13:14:07 -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

Tiny refactor weight loading logic (#5232)

2025-05-08 01:02:56 -07:00

models

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

2025-05-16 12:26:15 -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

Clean up imports (#5467)

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

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

Fix torch.compile cacheing (#5259)

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

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

[CI] Fix ci tests (#5769)

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

utils.py

Reduce MoE memory usage (#6147)

2025-05-15 09:38:28 -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