Glen Liu ad1b4e4728 [Feature] overlap LoRA weight loading with compute (#15512)

2026-01-19 10:43:17 +08:00

..

batch_invariant_ops

…

batch_overlap

…

checkpoint_engine

…

compilation

tiny refactor pcg split op registration (#16863)

2026-01-10 23:45:28 +08:00

configs

Use swa radix cache and memory pool for gpt-oss model (#17261)

2026-01-18 13:47:07 +08:00

connector

…

constrained

Fix grammar sync across TP ranks (#17100)

2026-01-15 18:38:01 -08:00

debug_utils

…

disaggregation

[Bugfix] Fix PD accuracy when MTP is not configured on the prefill node (#17212)

2026-01-18 23:37:58 +08:00

distributed

Add clear error message when OOM with symmetric memory (#17038)

2026-01-16 20:44:37 +08:00

dllm

[DLLM] Implement initial dynamic batching for diffusion LLM (#14883)

2026-01-17 16:48:15 +08:00

elastic_ep

…

entrypoints

Update flashinfer to 0.6.1 (#15551)

2026-01-17 00:48:30 +08:00

eplb

Re-introduce the unit test of test_mooncake_ep_small (#16019)

2026-01-12 17:01:24 +08:00

function_call

Fix parallel tool call parsing bug when tool parameters contain arrays (#16345)

2026-01-10 19:31:38 -08:00

grpc

[gRPC] Add GetLoads RPC for comprehensive load metrics (#17087)

2026-01-14 12:05:10 -08:00

hardware_backend/npu

[NPU]Support GPT-OSS for NPU (#14197)

2026-01-19 04:13:41 +08:00

layers

[GLM 4.7] Add RTX 6000 Pro aka sm120 (#17235)

2026-01-18 13:33:19 -08:00

lora

[Feature] overlap LoRA weight loading with compute (#15512)

2026-01-19 10:43:17 +08:00

managers

[Feature] overlap LoRA weight loading with compute (#15512)

2026-01-19 10:43:17 +08:00

mem_cache

[BUGFIX] fix radix cache memory consumption to avoid OOM (#17191)

2026-01-17 16:47:37 +08:00

metrics

Code clean up for fp8 quantization (#16982)

2026-01-13 12:38:39 -08:00

model_executor

[Feature] overlap LoRA weight loading with compute (#15512)

2026-01-19 10:43:17 +08:00

model_loader

ci: enable offline mode when local cache is complete to avoid HF Hub … (#16121)

2026-01-15 20:15:33 -08:00

models

[NPU]Support GPT-OSS for NPU (#14197)

2026-01-19 04:13:41 +08:00

multimodal

[VLM][Reland] Refactor load_mm_data to improve performance (#16152)

2026-01-18 14:11:17 +08:00

multiplex

…

parser

VLM: enhance VL embedding model with video input support and revise warm-up strategy (#16635)

2026-01-08 22:12:01 +08:00

sampling

Revert "feat: reduce constrained-decoding overhead in TP" (#16845)

2026-01-10 11:39:38 +08:00

speculative

[SPEC_V2] Enable cudagraph draft_extend for trtllm_mla_backend and Acclen Fix for DP under cudagraph mode (#16974)

2026-01-18 15:56:21 +08:00

tokenizer

…

tracing

…

utils

[Fix] flashinfer_trtllm intermediate_size assertion with Qwen3 + TP=8 (#16824)

2026-01-17 09:24:05 +08:00

weight_sync

…

constants.py

…

environ.py

[jit-kernel] Add CuTe DSL GDN Decode Kernel (#15631)

2026-01-18 12:54:36 -08:00

server_args_config_parser.py

[ConfigArgumentMerger] Improve ConfigArgumentMerger compatibility with external callers (#17051)

2026-01-16 15:32:42 +08:00

server_args.py

[Feature] overlap LoRA weight loading with compute (#15512)

2026-01-19 10:43:17 +08:00