Commits vllm-project/vllm

padded nvfp4 quant kernel

yewentao256 committed 12 hours ago

f2d01fc1

[ROCm] Restore fast top_k_per_row kernels for sparse MLA when topk_tokens=2048 (#42072)

frida-andersson committed 12 hours ago

Verified 4d67d3bd

[Bugfix] Fix SM121 (DGX Spark) exclusion from Marlin/CUTLASS FP8 paths (#35568)

blake-snc committed 13 hours ago

Verified 06d020bb

[LMCacheMPConnector] Prioritize importing the lmcache_mp_connector from lmcache (#42596)

chunxiaozheng committed 13 hours ago

Verified f45c2108

[ROCm] Widen AITER fused AR RMSNorm 1-stage gate (#42409)

akii96 committed 14 hours ago

Verified be7a03ea

[Model Runner v2] Support reload weights (sleep mode) (#42673)

yewentao256 committed 15 hours ago

Verified 6147c702

[Model Runner V2] FP32 gumbel sampling. (#41775)

PatchouliTIS committed 15 hours ago

Verified 01625966

[ROCm][MLA] FP8 ASM prefill for AITER dense MLA backend on gfx950 (#42509)

maeehart committed 15 hours ago

Verified 46a95815

[Perf] Set IR Op Priority Once at Worker Init (#42631)

BadrBasowid committed 15 hours ago

Verified fb5bd03f

[Bugfix] Fix DeepGEMM context lens contiguity in MLA indexer (#42135)

mmangkad committed 16 hours ago

Verified ee58665a

[Perf] Optimize MLA attention `_v_up_proj` bmm by removing additional copy (#42561)

yewentao256 committed 16 hours ago

Verified 491e8d85

[Model Runner V2] Fix kv_connector `pre_forward` order (#42676)

yewentao256 committed 16 hours ago

Verified af9616d8

[ROCm] Widen OAI Triton MoE capability range to include gfx12 (RDNA4) (#37826)

laudney committed 16 hours ago

Verified d792d993

[Feat][RL] IPC weight sync optimizations: multigpu support and chunked packed tensors (#37476)

hao-aaron committed 16 hours ago

Verified e0a45f14

[Bugfix] DFlash FP8 KV-Cache (#42692)

benchislett committed 17 hours ago

Verified 0fe75502

[Bugfix] Ensure embeding model compilation on CPU (#42709)

bigPYJ1151 committed 20 hours ago

Verified 95cfe102

gemma3 multi-gpu bug-fix (#42630)

pmaybank committed 22 hours ago

Verified 1dc3fe08

fix: propagate revision/code_revision pins to all artifact boundaries (#42616)

jperezdealgaba committed 22 hours ago

Verified d26a28ab

[ROCm][CI] Stage B gating (#42025)

AndreasKaratzas committed 22 hours ago

Verified d735968f

DeepSeekV4-Pro enable cuda graph full and piecewise mode (#42604)

bobofang11235 committed 23 hours ago

Verified ccde9540

[Entrypoints] Split the pooling offline API into PoolingOfflineMixin. (#42267)

noooop committed 23 hours ago

Verified 75fd68c7

[Core][DSV4] Skip caching SWA blocks that can never serve a prefix-cache hit (#42258)

ivanium committed 23 hours ago

Verified 4b364f81

[Misc] Make it simpler to replace out-of-tree layer classes with related LoRA layers. (#42306)

paulyu12 committed 1 day ago

Verified 31fa757c

[Deprecation] Remove old locations of `get_tokenizer` and `resolve_hf_chat_template` (#35024)

DarkLight1337 committed 1 day ago

Verified 2676ab1e

[Bugfix] Clarify CPU backend memory error messages reference shared flag (#42479)

daniel-devlab committed 1 day ago

Verified 27b85d20

Update Intel Xeon model list and vLLM Benchmark Suite BKMs (#42607)

louie-tsai committed 1 day ago

Verified e30f39c4

[Bugfix] Fix inverted condition causing thinking_token_budget to be silently ignored (#41674)

JasonKeyiL committed 1 day ago

Verified bf610c2f

[Model] Support InternS2 Preview (#42705)

Isotr0py committed 1 day ago

Verified faa4b76a

[CPU][RISC-V] Add RVV-optimized attention kernels for RISC-V Vector Extension (#40119)

lyd1992 committed 1 day ago

Verified f351455f

[Bugfix] Fix incorrect chat template format for Qwen3.5 (#42660)

DarkLight1337 committed 1 day ago

Verified 56434e86