Commit Graph

11116 Commits

Author SHA1 Message Date
Kangyan-Zhou
037c3982af Fix H200 CI by commenting out Warmup Weights and JIT Compilation (#15139) 2025-12-14 18:04:28 -08:00
fzyzcjy
62b3fdae43 Fix cache aware wrong routing caused by incorrect load tracking (#15101) 2025-12-15 10:03:54 +08:00
Yibo Cai
1cd0c3bfb7 Add cyb70289 to CI permissions (#14938) 2025-12-14 17:51:30 -08:00
Douglas Yang
2c899431f8 fix: adjusting frequency for ci failure monitor (#15134) 2025-12-14 17:49:57 -08:00
Xiaoyu Zhang
4513f549ee [diffusion] fix: fix default resolution 720p width from 1080 to 1280 (#15058) 2025-12-15 09:16:47 +08:00
YAMY
c96903074c [NSA] Fix NSA backend assertion error when running DeepSeek-V3.2 PP with radix-cache (#15086) 2025-12-14 17:13:18 -08:00
fzyzcjy
4449c17011 [model-gateway] fix circuit breaker metrics (#15099)
Co-authored-by: Simo Lin <linsimo.mark@gmail.com>
2025-12-14 15:43:04 -08:00
fzyzcjy
5ca962ce7f [model-gateway] extract circuit breaker state struct (#15098) 2025-12-14 15:40:23 -08:00
Praneth Paruchuri
bab20a849e [model-gateway] Parallelize metrics requests (#14953) 2025-12-14 15:38:45 -08:00
ratish
0e4108ba29 feat(gateway): Add server-side TLS support (#15052) 2025-12-14 15:37:55 -08:00
Hudson Xing
99cb2ed988 [refactor] Move trtllm_fp8_kv_kernel to triton_ops directory (#15044) 2025-12-14 15:22:37 -08:00
Simo Lin
0612175ce1 [model-gateway] add streaming metrics (TTFT, TPOT, tokens, duration) for gRPC router (#15125) 2025-12-14 14:39:10 -08:00
Douglas Yang
8c96fcda70 feature: ci failure monitor slack bot (#15110)
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 13:47:15 -08:00
zhangheng
ea7c69ce28 [hotfix]: Add missing args for 3FS bench_client.py (#14791)
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
2025-12-14 11:13:19 -10:00
danielafrimi
8102e36b5d Add NanoV3 reasoning parser support (#15113) 2025-12-14 12:09:44 -08:00
roikoren755
3f0482174a Fix Mamba2-based models' default attention backend (#15117)
Signed-off-by: Roi Koren <roik@nvidia.com>
2025-12-14 11:08:48 -08:00
Simo Lin
b11af135eb [model-gateway] feat(metrics): implement Layer 2 router metrics (smg_router_*) (#15124) 2025-12-14 10:34:12 -08:00
Simo Lin
f9bceea064 [model-gateway] Implement Layer 1 HTTP metrics instrumentation (#15121) 2025-12-14 09:39:15 -08:00
Ziming Huang
997ea57eaf Fix tensor mismatch error in sepc + topk > 1 + page_size > 1 (#14874) 2025-12-15 01:10:58 +08:00
Simo Lin
47633c191f [model-gateway] Add new SMG metrics architecture with 6 layers (#15106) 2025-12-14 08:37:19 -08:00
Xiaoyu Zhang
64b5c3ab90 [diffusion] refactor: refactor fuse qkv with QKVParallelLinear linear (#15090)
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-12-15 00:33:29 +08:00
Shangming Cai
d277a86dea [CI] Add disaggregation decode PP test (#15114) 2025-12-14 23:54:25 +08:00
Tianyu Guo
9acb21ae27 feat: support EPD disaggregation (#12263)
Co-authored-by: liusy58 <liusy58@linux.alibaba.com>
Co-authored-by: ZhengWG <zwg0606@gmail.com>
Co-authored-by: Nicholas <45984215+liusy58@users.noreply.github.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
Co-authored-by: Yuhao Yang <47235274+yhyang201@users.noreply.github.com>
2025-12-14 22:30:08 +08:00
zyl_keep_moving
a9ce1623cd [kernel][moe] add moe topk fast (#13969)
Co-authored-by: Xiaoyu Zhang <35585791+BBuf@users.noreply.github.com>
2025-12-14 22:26:40 +08:00
Li Jinliang
6f0c77d7f8 [diffusion] app: support webui (#14961)
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-12-14 21:07:36 +08:00
Xiaoyu Zhang
e3f51e823e [diffusion] feat: add support for additional sampling parameters in video generation API (#15062)
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-12-14 19:44:03 +08:00
Xiaoyu Zhang
fdfabb7afc [diffusion] fix: tiny fix _templated_ring_attention bug (#15053)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-12-14 19:41:53 +08:00
blahblah
19c16748ce [diffusion] feat: support resolution check for video model (#14881)
Co-authored-by: Brain97 <Brain97@users.noreply.github.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-12-14 17:50:13 +08:00
Liangsheng Yin
5c75907e62 Introduce native kv cache move (#15108) 2025-12-14 17:23:28 +08:00
liupeng374
4ea3642250 [NPU] bug fix for multi stream (#15048) 2025-12-14 16:38:12 +08:00
liupeng374
f50af32d8b [scheduler] remove scheduler allgather for best throughout (#14294) 2025-12-14 16:37:49 +08:00
Qiaolin Yu
729529190d [ci] Move dpsk-r1-fp4 b200 test to stage b (#15084) 2025-12-13 23:14:08 -08:00
monkeyLoveding
2ae5bed193 [NPU][CI] change de trigger of release image workflow (#14969) 2025-12-14 15:11:47 +08:00
fzyzcjy
54df514bac Avoid confusing zero value metric when worker is removed (#15096) 2025-12-14 13:57:20 +08:00
fzyzcjy
681c68cfbf Fix issue not reported when load decrement is incorrect (#15061)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-12-14 13:17:35 +08:00
Simo Lin
74ea45cc4b [model-gateway] optimize metric labels to avoid unnecessary allocations (#15095) 2025-12-13 21:10:45 -08:00
Simo Lin
0fa044ad63 [model-gateway] Add circuit breaker and discovery watcher metrics (#15094) 2025-12-13 20:47:04 -08:00
Simo Lin
7d8e42c979 [model-gateway] Fix metric emission gaps and name mismatch (#15093) 2025-12-13 20:16:30 -08:00
Huang Lin
6abdf73f4d [Fix] Environment variable SGL_* is deprecated (#14943) 2025-12-13 19:55:43 -08:00
Simo Lin
20ce9938b5 [model-gateway] Remove unused TokenizerMetrics to reduce CPU overhead (#15087) 2025-12-13 18:54:18 -08:00
fzyzcjy
168a31eb00 Support prefill max requests limitation (#14993) 2025-12-14 10:52:10 +08:00
narutolhy
69cfb17b9a [Fix] avoid stream sync in _compute_mrope_positions (#14956) 2025-12-13 18:50:57 -08:00
Brian
a7a4b1755d [Doc][TPU]add sglang-jax tpu docs (#15056) 2025-12-14 09:29:41 +08:00
fzyzcjy
fdc93b019c Add sglang:decode_sum_seq_lens metric (#15066) 2025-12-14 08:42:45 +08:00
Simo Lin
fd37cc5d38 [model-gateway] Refactor worker steps and add update workflow (#15085) 2025-12-13 15:00:16 -08:00
Xinyuan Tong
96705514bd fix: dpskv32 chat history processing, default drop_thinking to true (#15064)
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
2025-12-13 12:16:04 -08:00
Baizhou Zhang
ab3ffd1c8e Add nightly accuracy test for DeepSeek V3.2 (#14935) 2025-12-13 12:11:16 -08:00
Wenyi Xu
2285affffa [model-gateway] Avoid MCP Server Initialization Issue (#15065) 2025-12-13 11:20:16 -08:00
Yuhao Yang
a81cc1b8b3 add transformers version validation for glm-4.6v moe models (#14998) 2025-12-13 10:54:08 -08:00
liupeng374
3134d2b2a7 [scheduler] enhance scheduler in dp_attention mixed case with spec (#14201) 2025-12-14 02:52:26 +08:00