Lifu Huang 79a39ac0cc follow-up: move Idefics2 to a shared location to eliminate unexpected dependency. (#6603)

2025-05-26 19:23:59 -07:00

..

configs

chore: upgrade transformers 4.52.3 (#6575)

2025-05-25 22:49:58 -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

[PD] Handle P/D failure and reconnect without affecting other instances (#6263)

2025-05-26 19:21:01 -07:00

distributed

Use monotonic clock for interval measurement (#6211)

2025-05-17 16:49:18 -07:00

entrypoints

feat(Tool Calling): Support required and specific function mode (#6550)

2025-05-23 21:00:37 -07:00

function_call

feat: Improve Mistral and Qwen25 function call parsing (#6597)

2025-05-25 23:07:23 -07:00

layers

Tiny fix sampler error when prob is not contiguous (#6639)

2025-05-26 19:19:08 -07:00

lora

Fix lora bench (#6302)

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

managers

follow-up: move Idefics2 to a shared location to eliminate unexpected dependency. (#6603)

2025-05-26 19:23:59 -07:00

mem_cache

[VLM] Support chunk prefill for VLM (#6355)

2025-05-22 20:32:41 -07:00

metrics

Use monotonic clock for interval measurement (#6211)

2025-05-17 16:49:18 -07:00

model_executor

fix qwen3moe eplb prefill bug (#6617)

2025-05-26 02:15:21 -07:00

model_loader

Reland tiny refactor DefaultModelLoader.Source (#6041)

2025-05-17 17:11:20 -07:00

models

follow-up: move Idefics2 to a shared location to eliminate unexpected dependency. (#6603)

2025-05-26 19:23:59 -07:00

openai_api

feat(Tool Calling): Support required and specific function mode (#6550)

2025-05-23 21:00:37 -07:00

sampling

Skip the flaky test_stateful_custom_logit_processor (#6251)

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

speculative

Apply constraint grammar to EAGLE (#6499)

2025-05-21 17:18:41 -07:00

_custom_ops.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -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

Add typo checker in pre-commit (#6179)

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

conversation.py

Support Phi-4 Multi-Modal (text + vision only) (#6494)

2025-05-24 21:43:38 -07:00

custom_op.py

Clean up imports (#5467)

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

hf_transformers_utils.py

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00

mm_utils.py

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00

model_parallel.py

Make torch TP composable with torchao (#2436)

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

operations_strategy.py

qwen3moe support two batch overlap (#6598)

2025-05-25 23:08:16 -07:00

operations.py

Support overlapping two batches (#4068)

2025-05-24 17:39:07 -07:00

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

Automatically configure for EPLB-related args (#6628)

2025-05-26 08:42:49 -07:00

torch_memory_saver_adapter.py

[CI] Fix ci tests (#5769)

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

two_batch_overlap.py

qwen3moe support two batch overlap (#6598)

2025-05-25 23:08:16 -07:00

utils.py

refactor qwen moe code, use communicator to support tp+dp (#6581)

2025-05-25 23:01:10 -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