Commit Graph
17568 Commits
Author SHA1 Message Date
ce7e79b32c [AMD] Fix nightly ROCm 7.0 image build: patch missing <optional> include in AITER topk kernel (#36216)
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: quitenode <quitenode@users.noreply.github.com>
Co-authored-by: Chen <bingxche@amd.com>
Co-authored-by: YC Yen-Ching Tseng <yctseng@amd.com>
2026-09-01 19:06:32 +08:00
Eric.Chin.AMDandKingRei ed122ea984 [AMD] Enable topk v2 GLM ROCm (#36851)
Co-authored-by: KingRei <hiroki1139@gmail.com>
2026-09-01 03:48:50 -07:00
YC Yen-Ching Tseng b425897366 [AMD] Gate the aiter memory-reserve exemption behind an env var (#37242) 2026-09-01 03:46:15 -07:00
datdo-msftandShangming Cai 49db27528a fix(test): deflake zmq load-snapshot round-trip tests (#35787)
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-01 18:14:55 +08:00
Jialin Ouyang a77283fb02 [Rust] Rename mem-cache to sglang-radix-tree (#37290) 2026-09-01 03:03:44 -07:00
chx96642264 c16a8fc899 [NPU] [bugfix] Fix NPU MLA HiCache backup accessing missing data_ptrs. (#36813) 2026-09-01 17:31:11 +08:00
zijiexiaandClaude Opus 5 dc1ae02684 [Cookbook] Add the DFlash2 speculative option to GLM-5.3 (#37392)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-09-01 09:08:28 +00:00
eeecho b68702be99 [DSV4] hc-prenorm: fuse the combine step into a Triton kernel (#35118) 2026-09-01 02:04:05 -07:00
zijiexia 6c72b49a57 Revert "[AMD] Add GLM-5.3-Flash recipes for MI300X, MI325X, and MI355X (#36608)" (#37380) 2026-09-01 01:25:13 -07:00
Ziang Li 5edcd0a445 [FlashInfer V0.6.18] feat(dsv4): support --dsa-topk-backend flashinfer with fused top-k (#33237) 2026-09-01 01:18:10 -07:00
Liangsheng Yin 3484f7f836 [mem_cache] Add free_kv_row to release a request's kv row by row range (#36721) 2026-09-01 01:14:43 -07:00
Xiaoyu ZhangandCursor 1c3ad92438 [Diffusion] Fuse FLUX.2 ModelOpt FP8 producers and QKV packing (#37162)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 16:14:29 +08:00
zijiexiaandClaude Fable 5 379e33d87e [Cookbook] Add NVFP4 options for DeepSeek-V4 Flash Official (0731) and Pro Official (0813) (#37351)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-09-01 08:03:39 +00:00
Mohammad Miadh Angkad 03b33cbe5d [CI] Fix hybrid wrapper test fake missing kv_index_translator (#37374) 2026-09-01 00:55:05 -07:00
chien-an-chen 8a191554e3 [AMD] Enable 12-head MLA aiter fp8 Gluon decode (batched bh16bn128). (#34647) 2026-08-31 23:47:45 -07:00
Ma Mingfei 5e79110122 [Fix][CPU] fix xeon ci failure by test_qwen35_flashinfer_fusion (#37338) 2026-09-01 14:17:35 +08:00
Xinyuan Tong 60548501bb [Docs] Add NVFP4 section to GLM-5.3-Flash cookbook (#37109) 2026-09-01 14:14:30 +08:00
Liangsheng Yin 959ca033eb refactor(hicache): simplify decode offload state bookkeeping (#37299) 2026-08-31 23:09:41 -07:00
Mick ae2bd5728b [vlm] fix: contain multimodal feature transport failures (#37047) 2026-09-01 13:46:38 +08:00
YAMY 33ed29a0ee test: update hybrid attention runner fixtures (#37345) 2026-08-31 22:00:50 -07:00
Zhaoyi Li 458987b5ac [AMD][MORI] Bump MoRI to 7c51d18 for ionic RoCE dmabuf fix (#509) (#37286) 2026-08-31 21:37:05 -07:00
huangtingweiandhzh0425 b21000aef1 [Unified Cache][4/N]: Add Mooncake backend for external linker (#37205)
Co-authored-by: hzh0425 <hzh0425@apache.org>
2026-09-01 11:40:06 +08:00
Liangsheng Yin 3b14f37b74 [Fix] Use real ReqKvInfo in unit-test req mocks (#37339) 2026-08-31 20:25:51 -07:00
hhhh1252023 60f881b40c [CI/NPU] Isolate multi-node tests by run_id to prevent concurrent-run… (#35500) 2026-09-01 11:08:27 +08:00
Yuan Luoandluoyuan.luo 5b04408784 [MoE] Add FlashInfer SM90 MXFP4 W4A8 CUTLASS MoE (#34967)
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
2026-08-31 20:04:41 -07:00
22337e9c56 fix(unified-memory): forward the KV-index translator through every wrapper backend (#37307)
Co-authored-by: Caihua Li <caihua.li@bytedance.com>
Co-authored-by: Cheng Wan <cheng.wan@radixark.ai>
2026-08-31 19:55:21 -07:00
Mohammad Miadh Angkad e6f21cdadc [Cohere Command-A-Plus] Optimize decode and BCG capture on SM10X (#36624) 2026-08-31 19:37:47 -07:00
Yiqi Yang 1591dcd91a [diffusion] fix: fix loading a block-FP8 quantized MiniMax-H3 DiT (#35703) 2026-09-01 10:33:58 +08:00
71cee04ebe [Diffusion] Optimize Qwen-Image TP collectives and attention (#36680)
Co-authored-by: Mick <mickjagger19@icloud.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 10:28:37 +08:00
Martin Hua 562b661e0e [Feature] Megatron LayerNorm sequence parallelism (--enable-layernorm-sp) (#30915) 2026-08-31 19:27:28 -07:00
ashwini rathiandarathi-hlab 783af667fb xpu: record per-model metrics to jsonl for nightly dashboard (#36699)
Co-authored-by: arathi-hlab <arathi-hlab@users.noreply.github.com>
2026-09-01 10:20:56 +08:00
f50b4ad7ae [DCP] Support decode context parallelism on the trtllm_mla decode path (#33926)
Co-authored-by: kpham-sgl <khoa.pham@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-31 19:19:44 -07:00
Khoa Pham 97744189b8 [Kernel] Raise shape limits in shared FLA and MoE kernels (ported from #36507) (#37317) 2026-08-31 18:53:46 -07:00
Ziang Li 9a85473a89 [FlashInfer v0.6.18] add FlashInfer CuTe DSL NVFP4 W4A16 mode (#35120) 2026-08-31 18:47:30 -07:00
Xiaoyu ZhangandCursor 175973d834 [Diffusion] Fuse Qwen-Image residual norm and NVFP4 quantization (#37129)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 08:54:10 +08:00
Xiaoyu ZhangandCursor 079afaffb1 [Diffusion] Fuse FLUX.2 gated residual normalization on Blackwell (#37112)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 08:38:33 +08:00
Xiaoyu ZhangandCursor 6715debb2a [Diffusion] Fuse Qwen-Image FP8 QKV projection and Blackwell epilogue (#37123)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 08:34:41 +08:00
Baizhou Zhang 07c8f7294d Bump sgl-deep-gemm to 0.1.7 (#37279) 2026-08-31 17:16:34 -07:00
zijiexia 455232de6e [Cookbook] Enable DSpark on the DeepSeek-V4 Flash Vision low-latency recipes (#37301) 2026-08-31 16:43:54 -07:00
ef9e58fd6d feat(unified-memory): three sub-pools for mamba + hybrid-SWA models (#35177)
Co-authored-by: Caihua Li <caihua.li@bytedance.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Cheng Wan <cheng.wan@radixark.ai>
2026-08-31 15:10:12 -07:00
98cb3535b7 feat(unified-memory): byte-budget sizing, feasibility floor, and a conservation verifier (#35158)
Co-authored-by: Caihua Li <caihua.li@bytedance.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Cheng Wan <cheng.wan@radixark.ai>
2026-08-31 15:09:28 -07:00
961beee9e5 fix(unified-memory): four boot/correctness fixes on the hybrid model paths (#35154)
Co-authored-by: Caihua Li <caihua.li@bytedance.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Cheng Wan <cheng.wan@radixark.ai>
2026-08-31 15:08:43 -07:00
zijiexia 88cf5c9541 [Cookbook] Add DeepSeek-V4-Flash-Vision-Exp to the DeepSeek-V4 page (#37293) 2026-08-31 14:52:28 -07:00
Liangsheng Yin 2530204502 [mem_cache] Make release, row-reuse asserts, and presence checks read the KV record (#37167) 2026-08-31 12:46:15 -07:00
Liangsheng YinandBaizhou Zhang 95f0f41021 [CI] Move tests onto the right CI stages (#34074)
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
2026-08-31 12:39:59 -07:00
Lianmin Zheng 579270d459 [Rust] Simplify request defaults and document batch header ABI (#37226) 2026-08-31 12:31:35 -07:00
Lianmin Zheng 2138494272 [Rust] Keep sampling and scheduler wire schemas in sync (#37222) 2026-08-31 12:31:07 -07:00
Lianmin Zheng 48098b5f23 [Rust] Derive server address and accept signed env values (#37221) 2026-08-31 12:30:40 -07:00
Lianmin Zheng 1da86b9801 [Rust] Split and rename embedded server components (#37220) 2026-08-31 12:28:43 -07:00
ishandhanani cf51650335 fix(config): retain pre-engine resolution declarations (#37195)
Signed-off-by: Ishan Dhanani <ishandhanani@gmail.com>
2026-08-31 12:24:29 -07:00