Commit Graph

Select branches

Hide Pull Requests

bot/update-est-time-20260928

dsv41-pd

dsv41-pd-visioncp

main

Mono Color

• 6e3fff1352 fix: adding retries for server start and increasing hf read timeout (#16523) Douglas Yang 2026-01-06 09:53:12 -08:00
• 2210155a45 ci: migrate DLLM tests to test/registered/dllm/ (#16420) Alison Shao 2026-01-06 09:44:26 -08:00
• a3656cbb96 [ci] fix url strips in smg ci (#16548) Simo LinandChang Su 2026-01-06 09:23:58 -08:00
• 21da2dc1ce [grpc] Unify ResponsesContext and HarmonyResponsesContext (#16549) Chang Su 2026-01-06 07:28:06 -08:00
• ed307a40bf Tiny add --gsp-ordered (#16575) fzyzcjy 2026-01-06 22:43:52 +08:00
• 02722b9113 Fix evict swa for overlap scheduler and page size > 1 (#16507) Ke Bao 2026-01-06 22:29:40 +08:00
• f959250f76 Tiny support sglang_routing_keys_active in engine (#16570) fzyzcjy 2026-01-06 22:10:10 +08:00
• 959343795f [test] update acc len threshold to 2.7 for eagle dp attention tests (#16463) YAMY 2026-01-06 01:38:35 -08:00
• bc2f40bebc [test] Add mamba cache release/resume memory test (#14215) Junrong Lin 2026-01-06 16:51:10 +09:00
• 2724b1100b fix double Unicode escape issue in streaming tool_calls parameters (#13518) lw9527andXinyuan Tong 2026-01-06 15:47:26 +08:00
• 176266f358 Migrate sampling tests to test/registered/sampling/ (#16455) Alison Shao 2026-01-05 23:43:20 -08:00
• 5e5b1183ed [Diffusion] Ring Attention support sage backend (#16496) Xiaoyu Zhang 2026-01-06 14:45:53 +08:00
• 9bf76c11a9 [responses API] Add list_tools_for_servers and threading server_keys in routers (#16540) Chang Su 2026-01-05 22:42:22 -08:00
• 1d7ad4afcc [diffusion] fix: fix bench_serving always use prompt fixed prompts (#16201) 3 people 2026-01-06 14:39:07 +08:00
• fba785c459 use aync load for encoder_server (#15456) 3 people 2026-01-06 14:31:52 +08:00
• 5cfa901b57 ci: migrate Constrained Decoding tests to test/registered/constrained/ (#16421) Alison Shao 2026-01-05 22:29:22 -08:00
• f27c6cdc2b Migrate metrics tests to test/registered/metrics/ (#16466) Alison Shao 2026-01-05 22:21:49 -08:00
• 5097e1e85b fix: moving dpsk v32 cp single node to nightly (#16539) Douglas Yang 2026-01-05 22:18:59 -08:00
• 861a35fb6e ci: migrate RL tests to test/registered/rl/ (#16417) Alison Shao 2026-01-05 22:13:13 -08:00
• 6ffe1fc02f [Fix]Pin mooncake version to 0.3.7.post2 in grace blackwell (#16502) Baizhou Zhang 2026-01-06 14:11:40 +08:00
• 73398e22d6 ci: migrate VLM tests to test/registered/vlm/ (#16415) Alison Shao 2026-01-05 21:25:29 -08:00
• 17958c5f0d Migrate parsing tests to test/registered/parser/ (#16467) Alison Shao 2026-01-05 21:15:32 -08:00
• 3aa11ca722 Remove hybrid_kvcache_ratio in server args (#16399) Ke Bao 2026-01-06 13:13:13 +08:00
• 7be1a8c70c [diffusion] feat: support diffusers backend - run any model supported by diffusers (#14112) Adarsh Shirawalmath 2026-01-06 10:00:57 +05:30
• 84d13c54bb feat: only add input vision tokens in bench_serving result if vision dataset is used (#15492) Raayan Dhar 2026-01-05 20:14:42 -08:00
• c80c0e0fc0 fix: adding multi-threading for kimi weights loading in pr test (#16538) Douglas Yang 2026-01-05 20:10:13 -08:00
• c105a3124b Support multi-round conversations in bench_serving (#6135) fzyzcjy 2026-01-06 11:59:39 +08:00
• 4cf2bbd084 [router] Remove deadcode and add note for unused API completeness methods (#16528) Chang Su 2026-01-05 19:53:28 -08:00
• d7b706be94 ci: migrate HiCache 1-GPU tests to test/registered/hicache/ (#16416) Alison Shao 2026-01-05 19:38:21 -08:00
• 4a9537a495 ci: migrate Mamba/Layers tests to test/registered/layers/mamba/ (#16419) Alison Shao 2026-01-05 19:12:47 -08:00
• ca922d4b05 [diffusion] feat: support warmup with resolutions (#16434) Mick 2026-01-06 10:32:18 +08:00
• 402a0bd6dc [model-gateway] Add model scope support and LRU eviction for GPU-constrained environments (#16525) Simo Lin 2026-01-05 18:28:07 -08:00
• 76c71d1d34 fix: unimplemented methods in BaseIndexerMetadata (#16520) Douglas Yang 2026-01-05 18:26:42 -08:00
• 2d02c150dc [3/N][Sparse With Hicache]: Init sparse coordinator (#16086) 3 people 2026-01-06 09:51:04 +08:00
• b98bd9a5fb [model-gateway] Tighten visibility in modules and remove unused re-exports (#16524) Chang Su 2026-01-05 17:38:50 -08:00
• ce694b2b9e Fix the problem where Qwen3VL raises an "object has no attribute 'mod… (#15677) ShirakSyouya 2026-01-06 08:45:13 +08:00
• 6c0fb189f8 [Auto Sync] Update tokenizer_manager.py (20260105) (#16477) 3 people 2026-01-05 14:06:03 -08:00
• 9a9f996f8d [FP8] Fix weight_scale shape to match with x_scale shape for per-tensor quant under torch.compile (#16356) Binyao Jiang 2026-01-05 13:55:26 -08:00
• 4221b7c573 [model-gateway] refactor e2e test infrastructure and add router CI (#16513) Simo Lin 2026-01-05 13:50:59 -08:00
• c371df2f25 [AMD] Fix CI and add retry logic for git clone timeout (#15663) yctseng0211andBingxu Chen 2026-01-06 04:37:52 +08:00
• 1751c75b5d [model-gateway] Tighten visibility across data_connector and grpc module (#16516) Chang Su 2026-01-05 12:31:32 -08:00
• 23849eba7b [model-gateway] fix tokenizer encode in golang bindings (#16482) realWeilai 2026-01-06 03:40:09 +08:00
• 51541404f8 [grpc] Refactor openai module (#16511) Chang Su 2026-01-05 11:39:33 -08:00
• 45ef834412 Add MoE Integration Tests For CUTLASS Coverage (#16280) Jonah Bernard 2026-01-05 11:26:11 -08:00
• 5a2b1ed407 [grpc] Refactor grpc/regular/responses (#16509) Chang Su 2026-01-05 10:21:34 -08:00
• 454dc9e242 [model-gateway][grpc] Refactor harmony/responses.rs (#16508) Chang Su 2026-01-05 10:21:05 -08:00
• 1e41069ad1 Fix age bucket rendering issue (#16492) fzyzcjy 2026-01-06 01:29:14 +08:00
• 2b4d6d813b [model-gateway][e2e_test]: Create directory structure and backends config (#16469) Simo Lin 2026-01-05 08:01:53 -08:00
• a3914e3b3f [Bugfix] fix some memory computation bugs for qwen3next with mtp (#16138) Yi Zhang 2026-01-05 23:24:21 +08:00
• 130f60ee78 Add CODEOWNERS for scheduler_pp_mixin.py (#16481) Shangming Cai 2026-01-05 21:12:20 +08:00
• 4397cda7dc feat: raise error in PD when page sizes are mismatched (#14474) Raayan DharandShangming Cai 2026-01-05 03:54:01 -08:00
• d56fd10cb6 [AMD CI] fix sglang version detection in daily docker image release and init container (#16367) Bingxu Chen 2026-01-05 19:22:39 +08:00
• abb06be946 [NPU] Fixed link redirection issue. (#16475) Hexq0210 2026-01-05 19:03:29 +08:00
• c35eb0fdee [Auto Sync] Update loader.py (20260105) (#16478) 3 people 2026-01-05 01:34:30 -08:00
• 7f35c46efb Tiny add sglang:http_requests_active metric (#16479) fzyzcjy 2026-01-05 17:20:31 +08:00
• da2f8cc33f [minor] reduce 1 unnecessary add (#16474) DarkSharpness 2026-01-05 17:00:56 +08:00
• 4308c25b47 Super tiny move test folder (#16447) fzyzcjy 2026-01-05 16:29:26 +08:00
• 4d737db857 Tiny fix prefill delayer not support non-fcfs schedule policy (#16471) fzyzcjy 2026-01-05 16:28:54 +08:00
• 9d6029fb92 Fix TokenizerManager bottleneck for offline generation (#16456) fzyzcjy 2026-01-05 16:28:25 +08:00
• dcfb92ddbc [diffusion] fix: fix using the sub-optimal fa kernel (#16382) Qi Yuhang 2026-01-05 16:25:41 +08:00
• bebd625ba1 EVS Framework: Support NemotronH_Nano_VL_V2 (#14051) Netanel Haber 2026-01-05 10:18:07 +02:00
• b12258bfaa [model-gateway] Optimize HTTP Router Fan-out: Replace Serial Execution with Concurrent Streams (#16042) Praneth Paruchuri 2026-01-05 13:08:45 +05:30
• 012dc5866d [Auto Sync] Update scheduler.py (20260104) (#16424) 5 people 2026-01-04 23:32:56 -08:00
• 7f6a678f8f [model-gateway] add GPU allocator and model pool infrastructure for parallel E2E tests (#16460) Simo Lin 2026-01-04 23:30:32 -08:00
• 399ca037b1 [bug] fix lint blocking ci (#16464) Simo Lin 2026-01-04 22:57:33 -08:00
• d93f37a625 Add harvenstar to CI_PERMISSIONS.json Kangyan-Zhou 2026-01-04 22:35:26 -08:00
• e6fe092dcc [model-gateway] rename py_test to e2e_test (#16454) Simo Lin 2026-01-04 22:24:34 -08:00
• f02d82211a [model-gateway]: move PD configuration conflict checks to model gateway (#16088) Ratish P 2026-01-05 10:16:07 +04:00
• 10174e1114 Revert "[grpc] update api to scheduler in grpc request manager" (#16387) Yuhao Yang 2026-01-05 14:05:39 +08:00
• 2138ff48c6 Revert "[FEAT] optimize tensor zmq transfer for multimodal inputs" (#16386) Yuhao Yang 2026-01-05 14:05:26 +08:00
• e53160bb31 [model-gateway] reorganize integration tests into logical subdirectories (#16451) Simo Lin 2026-01-04 21:49:20 -08:00
• 4ea6a11c83 [CI] Fail wheel build when sgl-kernel artifacts are missing (#16450) Xiaoyu Zhang 2026-01-05 13:46:40 +08:00
• 2181bc9e51 [model-gateway] Delete Python integration_mock tests (#16448) Simo Lin 2026-01-04 21:34:06 -08:00
• 520c048d55 [diffusion] CI: add script for automatically generation ci perf baseline (#16389) Xiaoyu Zhang 2026-01-05 13:18:35 +08:00
• 561a3e04f6 remove redundant mamba_cache clearing actions (#16180) yudian0504 2026-01-05 13:02:07 +08:00
• f84487af59 [model-gateway] : Rust integration tests for integration_mock replacement (#16441) Simo Lin 2026-01-04 21:01:47 -08:00
• 078270473a [Doc] Default lora backend: csgmv (#16444) Huapeng Zhou 2026-01-04 20:45:49 -08:00
• c63e9cb29e Super tiny fix CI (#16437) fzyzcjy 2026-01-05 12:34:49 +08:00
• 12cde0df99 [SPEC_V2] Fix Acclen drop when enabling DP Attention for Spec-Overlap (#16310) YAMY 2026-01-04 19:39:23 -08:00
• a7fd810842 Allow editable install without .git with add fallback version in pyproject.toml (#16435) Liangsheng Yin 2026-01-05 11:17:20 +08:00
• ca80c19b55 Revert "[diffusion] feat: support warmup with resolutions" (#16433) Kangyan-Zhou 2026-01-04 18:44:05 -08:00
• 1e7b326482 Super tiny fix main code (#16432) fzyzcjy 2026-01-05 10:34:25 +08:00
• e267ca0beb [diffusion] doc: document LoRA support in CLI (#16375) CHEN Xi 2026-01-05 10:20:30 +08:00
• 9a8ba3c189 [diffusion] feat: support warmup with resolutions (#16330) Mick 2026-01-05 10:16:26 +08:00
• 0fee6bc632 [JIT kernel] Apply jit per_tensor_quant_fp8 kernel (#15836) Xiaoyu Zhang 2026-01-05 10:15:00 +08:00
• 0ff3747ca1 [model-gateway]: move unit tests to bindings/python/tests/ (#16430) Simo Lin 2026-01-04 18:10:35 -08:00
• f8411ded6e ci: migrate 1-GPU model tests to test/registered/models/ (#16414) Alison Shao 2026-01-04 18:08:01 -08:00
• 249c356331 Super tiny update tokenizer benchmark (#16429) fzyzcjy 2026-01-05 09:14:52 +08:00
• 12df16607b Tiny speed up kimi detokenizer by 10x (#16427) fzyzcjy 2026-01-05 09:12:05 +08:00
• 55d112dc79 fix: enable multi-threading for h200 tests (#16413) Douglas Yang 2026-01-04 14:43:53 -08:00
• a1ed247fd7 feature: add runner online count to failure monitor (#16408) Douglas Yang 2026-01-04 13:24:04 -08:00
• 87699d48eb fix: only publish trace from tp 0 (#16411) Douglas Yang 2026-01-04 12:17:08 -08:00
• 52c604342c chore: print test list at beginning and end of run_suite.py (#16334) Alison ShaoandMick 2026-01-04 11:52:52 -08:00
• ff0f370f85 ci: migrate MoE tests to test/registered/moe/ (#16127) Alison Shao 2026-01-04 11:51:12 -08:00
• 26f9e20755 ci: migrate quantization kernel tests to test/registered/quant/ (#16323) Alison Shao 2026-01-04 11:49:29 -08:00
• 828cd8936f Introduce sgl-kernel Dockerfile (#14066) Yingchun Lai 2026-01-05 03:19:08 +08:00
• 4436dc0f6c [model-gateway] improve lock contention and allocation in middleware (#16405) Simo Lin 2026-01-04 10:40:07 -08:00
• cf6800f6cc [model-gateway] optimize Vec and HashMap allocations in responses api (#16406) Simo Lin 2026-01-04 10:07:55 -08:00
• f16606d6f6 Super tiny code cleanup (#16401) fzyzcjy 2026-01-04 23:00:09 +08:00
• 387fad2f74 Tiny add detokenization benchmarks (#16400) fzyzcjy 2026-01-04 22:53:38 +08:00