A high-throughput and memory-efficient inference and serving engine for LLMs
main
[Refactor] Remove dead code quantization 2 (#52821)
Wentao Ye committed about 2 months ago
[SM100] Hdim 256 optimized (#52980)
simon-veitner-redhat committed about 2 months ago
[Bugfix][LoRA] Enable tower/connector LoRA for Qwen3-Omni (#53557)
linitra24 committed about 2 months ago
[Perf] Tune FlashInfer all-reduce thresholds for single-node TP8 on SM103 (#53606)
Jee Jee Li committed about 2 months ago
[Agents] Add CUDA IMA debugging skill (#53702)
Thien Tran committed about 2 months ago
Wentao Ye committed about 2 months ago
[Kimi K3][Kernel] Enable low-latency decode GEMM dispatch on SM100 (#53534)
Canlin Guo committed about 2 months ago
[Bugfix][KV Offload] Defer request-level cascade of in-flight primary keys (#53329)
Almog Tavor committed about 2 months ago
[Core] Disable fuse_allreduce_rms under VLLM_BATCH_INVARIANT (non-deterministic under TP) (#51292)
Donald Tolley committed about 2 months ago
[CI Bug] Fix kimi test `AssertionError: Aligned Mamba state indices must be precomputed` (#53766)
Wentao Ye committed about 2 months ago
[Perf] Autotune batch invariance triton kernel in blackwell, 33.6% E2E latency reduction (#53649)
Wentao Ye committed about 2 months ago
[Bugfix][MRV2] Run cudagraph memory profiling in a throwaway graph pool (#53682)
Nick Hill committed about 2 months ago
[CI] Preserve Rust Docker cache across commits (#53290)
Misha Goin committed about 2 months ago
[Bugfix][Multimodal] Reject malformed base64 audio with 400 instead of 500 (#53744)
Moe Huzaifa committed about 2 months ago
[Spec Decode] Enable adaptive DSpark on SM100 sparse MLA (#52783)
Misha Goin committed about 2 months ago
[Mypy Fix] Mypy fix for "vllm/model_executor/models/[gG]" (#53616)
Wentao Ye committed about 2 months ago
[Bugfix][Frontend] Apply the stop string limit to Cohere requests (#53750)
Moe Huzaifa committed about 2 months ago
[Bugfix][Frontend] Keep credentials out of the Rust frontend launch log (#53738)
Moe Huzaifa committed about 2 months ago
[Bugfix][Tokenizer] Replace bare asserts in the DeepSeek V4 encoder (#53747)
Moe Huzaifa committed about 2 months ago
[Bugfix] Resolve B12X modules before Dynamo tracing (#53326)
Luke Alonso committed about 2 months ago