Commit Graph
13715 Commits
Author SHA1 Message Date
Yaochen Hanandronnie_zheng 1f5dc2cdca [GPTQ] Refactor CPU quantization schemes (#26786)
Co-authored-by: ronnie_zheng <zl19940307@163.com>
2026-06-08 13:14:40 +03:00
YC Yen-Ching Tseng a26587dd4e [AMD][diffusion] Add FlyDSL fused normalization kernels for ROCm diffusion models optimization (#22786) 2026-06-08 02:42:39 -07:00
Liangsheng Yin 57ea09badb Fix NaN in triton EAGLE spec-v2 draft-extend CUDA graph at topk>1 (wrong qo_indptr stride) (#27545) 2026-06-08 02:28:48 -07:00
Shangming Cai 8ff0c9fef9 [PD] Downgrade propagated rank failure logs from error to debug (#27534)
Signed-off-by: Shangming Cai <csmthu@gmail.com>
2026-06-08 16:56:46 +08:00
Bingxu ChenandCursor df6b9c2d9d [AMD] ci: reinstall MoRI if Different from Dockerfile-pinned commit during install_dependency (#27538)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-08 16:29:01 +08:00
Ke Bao d03182cd2d Fix TP deadlock in unified radix cache writing_check / loading_check (#27489) 2026-06-08 16:08:47 +08:00
siyu 13dda3b8de [EPD] Support dynamic encoder register (#22253) 2026-06-08 16:05:16 +08:00
Niko Ma 18d728967a [PD][MoRI] Drive KV transfers with a sharded synchronous worker pool (#26922) 2026-06-08 00:49:25 -07:00
Jan BernlöhrandClaude Opus 4.6 0d0254c9de Fix port overflow in DP attention path when base port is near 65535 (#20260)
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-06-08 00:39:29 -07:00
fzyzcjy 71a0b10462 Fix the _chunked_req_scheduled_last_iter flag with a content-based stash gate (#26938) 2026-06-08 14:55:41 +08:00
fzyzcjy f746e4a608 Fix fill_len asymmetric assignment statement in ignore-eos branch (#26999) 2026-06-08 14:53:21 +08:00
fzyzcjy 9034c2f9ae Fix Req fill_len (fill_ids) having dual semantics by restricting to truncated/committed semantics (#26659) 2026-06-08 14:52:48 +08:00
fzyzcjy 259a2da3e0 Refactor Req.fill_ids into full_untruncated_fill_ids + fill_len with equivalence (#26637) 2026-06-08 14:52:18 +08:00
fzyzcjy 4201de11de Extract release_req and retract_all as module-level free functions (#26548) 2026-06-08 14:51:45 +08:00
fzyzcjy 3197808283 Avoid calling filter_batch with chunked_req_to_exclude being things unrelated to chunked reqs (#26547) 2026-06-08 14:51:21 +08:00
fzyzcjy f5fdf9c5d8 Speed up dump comparator percentile computation using numpy (#26874) 2026-06-08 14:50:31 +08:00
fzyzcjy 995e649190 Add parallel-rank dump filenames and pipeline-global layer remapping to dumper (#26850) 2026-06-08 14:49:22 +08:00
Han Yu 3d2165a286 Fix dual-chunk sparse fallback index overflow (#27361) 2026-06-07 23:15:37 -07:00
HAI 1aa5040c74 Update code owners (AMD) (#27530) 2026-06-07 22:43:51 -07:00
zijiexiaandClaude Opus 4.8 d1777d1f6d Cookbook renovation (#26885)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-07 22:04:54 -07:00
Liangsheng Yin 6365d6faee [spec] Misc defensive guards for EAGLE draft KV indexing (#27486) 2026-06-07 21:27:46 -07:00
Xiaoyu Zhangandxiaoyu.zhang 6c2770149b [diffusion] Fix native text-encoder loading for T5/UMT5 encoder-decoder models (#27432)
Co-authored-by: xiaoyu.zhang <xiaoyu.zhang@radixark.net>
2026-06-08 12:23:16 +08:00
Xiaoyu Zhang 5bf7dd8e4a Update SGLang diffusion skills (#27496) 2026-06-08 12:22:45 +08:00
David Wang bf7fb6b925 fix dflash rope config parsing for updated transformers (#27477) 2026-06-07 20:52:07 -07:00
ashwini rathiandClaude Opus 4.7 acf65c1c68 [XPU CI] Pull prebuilt nightly image instead of building per-stage (#27282)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-06-08 11:08:41 +08:00
kkandwunhuang 1c73ff8ad3 [AMD] Optimize gpt-oss-120B performance (#27063)
Co-authored-by: wunhuang <wunhuang@amd.com>
2026-06-07 20:06:30 -07:00
Kangyan-ZhouandClaude Opus 4.8 6bc3953b48 feat(agentic router): add sticky-session routing policy (#27394)
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-07 20:05:27 -07:00
Jitendra Patil 12d47fa78b ci(xpu): push latest tag and use 7-char SHA in nightly image tags (#27299) 2026-06-08 10:49:54 +08:00
Cao E 2d1856bf45 Support encoder_decoder on cpu_graph_runner (#10950) 2026-06-08 10:30:11 +08:00
Liangsheng Yin 303757ccd8 [Attn] Fix aiter MLA verify kv_indices under-alloc + shared assert_buffer_fits guard (#27485) 2026-06-07 17:06:25 -07:00
Liangsheng Yin f68c79675f Support topk > 1 tree drafting for mamba/hybrid-linear models on spec v2 (#27463) 2026-06-07 17:04:09 -07:00
Liangsheng Yin 70db73afce [Spec] Clamp multimodal pad sentinels in spec-v2 draft prefill embedding (#27512) 2026-06-07 17:00:49 -07:00
Hubert Lu 10d33bd77e [AMD] Enable Piecewise CUDA Graph for AMD GPUs (#22299) 2026-06-07 16:28:20 -07:00
Mick 02be2e7189 [diffusion] support tp for ideogram4 (#27393) 2026-06-07 23:14:58 +08:00
Lianmin Zheng a07d813ec8 Revert "Fix TRTLLM target verify query metadata (#27473)" (#27494) 2026-06-07 02:48:52 -07:00
Lianmin ZhengandMing Yang db58e76c33 Add all_to_all_single to GroupCoordinator (#27492)
Co-authored-by: Ming Yang <minos.future@gmail.com>
2026-06-07 02:48:47 -07:00
Lianmin Zheng 5be0b0c8c0 Fix TRTLLM target verify query metadata (#27473) 2026-06-07 02:46:01 -07:00
fzyzcjy 0a190d1c97 Fix PP is_fully_idle missing in-flight microbatches (#27446) 2026-06-07 17:38:26 +08:00
fzyzcjy 14b8f98a21 Complete server warmup before scripted runtime scripts start (#27445) 2026-06-07 17:37:37 +08:00
Zhangheng a39c428d3f [UnifiedTree][CI]: Reduce HiCache PP KL test concurrency to avoid decode OOM (#27483) 2026-06-07 17:12:42 +08:00
Liangsheng Yin eab2e02fa0 [spec] Dedup draft kv_indices sizing into spec_utils helpers (#27475) 2026-06-07 00:42:29 -07:00
Liangsheng Yin 5e2e0d5b49 [spec] Make spec_utils module-importable: type-only imports under TYPE_CHECKING (#27484) 2026-06-07 00:01:32 -07:00
Xiaoyu Zhang ff8b97406d [diffusion] optimize: precompute ideogram4 denoising metadata (#27443) 2026-06-07 14:27:41 +08:00
Liangsheng Yin 0ce3db3c0a [Bug] Fix out-of-range token id crashing tp=1 VocabParallelEmbedding (#27482) 2026-06-06 23:00:17 -07:00
Liangsheng Yin 80eee2d92c [Spec] Guard async-assert probes against None tensor (#27478) 2026-06-06 22:14:00 -07:00
52a5c01eba [plugin] enable OOT platforms to provide custom quant configs (#25347)
Signed-off-by: Devashish Lal <devcode@fb.com>
Co-authored-by: Devashish Lal <devcode@fb.com>
Co-authored-by: Devashish Lal <laldevashish@gmail.com>
2026-06-07 12:48:55 +08:00
huangtingwei 857ecb2dbc [HiSparse & HiCache]Support mooncake store layer first layout (#27454) 2026-06-07 12:46:42 +08:00
Mohammad Miadh AngkadandLianmin Zheng 52f221cce0 Fix Req array token-id concatenation (#26182)
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
2026-06-06 19:59:51 -07:00
Liangsheng Yin 066b4a2180 [core] Probe set_kv_buffer / set_mla_kv_buffer slot ids for OOB (#27459) 2026-06-06 19:51:09 -07:00
Zhangheng fe548f36b0 [UnifiedTree]: Fix SWA admission budget under-counts HiCache load-back consumption (#27391) 2026-06-07 10:47:08 +08:00