Zhiqiang Xie 675acecec6 Attention backend selection bug fix for hicache (#16779) ...

Co-authored-by: hxie <haiyong.xie@ustc.edu>

2026-01-10 00:22:41 -08:00

..

batch_invariant_ops

…

batch_overlap

[Feature] Xiaomi MiMo-V2-Flash day0 support (#15207)

2025-12-19 11:40:07 +08:00

checkpoint_engine

…

compilation

Custom All Reduce for Piecewise Cuda Graph (#15356)

2025-12-25 23:55:15 +08:00

configs

Remove hybrid_kvcache_ratio in server args (#16399)

2026-01-06 13:13:13 +08:00

connector

…

constrained

Adding tool calling and reasoning parser support for Intern-S1 (#14866)

2025-12-16 00:01:58 -08:00

debug_utils

Support multiple engines for router simulation in schedule simulator (#16368)

2026-01-04 11:46:05 +08:00

disaggregation

[HiCacheStorage & PD] fix prefill bootstrap request host memory leaks (#15439)

2026-01-07 11:03:57 +08:00

distributed

ci: migrate 2-GPU tests to test/registered/ (#16529)

2026-01-07 20:28:16 -08:00

dllm

Tiny extract ModelRunnerOutput (#15400)

2025-12-18 22:18:45 +08:00

elastic_ep

…

entrypoints

Tiny pass routing key to scheduler processes (#16839)

2026-01-10 10:13:00 +08:00

eplb

Support EPLB balancedness prometheus metric without GPU->CPU synchronize (#15401)

2025-12-18 22:24:23 +08:00

function_call

Fix GLM-4.7 MoE Detector complex JSON Schema type parsing (#15753)

2026-01-10 01:21:10 +08:00

grpc

[grpc] Auto-generate protobuf files during wheel build (#16409)

2026-01-08 09:09:54 -08:00

hardware_backend/npu

[NPU][Bugfix] move free_page logics to cpu (#16608)

2026-01-08 12:00:18 +08:00

layers

[Rework] Add SwapAB Optimization for triton fused_moe_kernel on SM90. (#16723)

2026-01-10 13:57:44 +08:00

lora

[LoRA] Torch native backend: rework implementation and updated tests (#15187)

2025-12-30 11:48:51 +08:00

managers

Tiny let soft watchdog cover initialization phase (#16853)

2026-01-10 13:20:08 +08:00

mem_cache

fix spec qwen3 pd error (#16708)

2026-01-09 19:20:17 +08:00

metrics

Tiny add CPU resource monitoring for overload diagnosis (#16852)

2026-01-10 12:48:18 +08:00

model_executor

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

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

model_loader

Support bitwise weight checksum verifier (#16729)

2026-01-08 20:19:25 +08:00

models

Fix EPLB + FP4 Quantization Compatibility Issue (#13715)

2026-01-10 13:38:19 +08:00

multimodal

[VLM] Fix CUDA IPC OOM (#16118)

2026-01-07 11:30:35 +08:00

multiplex

Fix deprecation warning of sre_parse for python 3.13 (#16247)

2025-12-31 21:25:09 -08:00

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

Tiny simplify draft worker init. (#16446)

2026-01-09 11:20:29 +08:00

tokenizer

…

tracing

SGLang Tracing: fix attribute errors (header extraction & bootstrap span closing) (#15693)

2025-12-27 22:44:31 -06:00

utils

Tiny add gauge histogram abstraction for engine and router (#16848)

2026-01-10 11:45:25 +08:00

weight_sync

…

constants.py

…

environ.py

Tiny let soft watchdog cover initialization phase (#16853)

2026-01-10 13:20:08 +08:00

server_args_config_parser.py

Fix parse args from file(#13911) (#14085)

2026-01-01 11:37:33 +08:00

server_args.py

Attention backend selection bug fix for hicache (#16779)

2026-01-10 00:22:41 -08:00