A high-throughput and memory-efficient inference and serving engine for LLMs
main
[Bugfix][Core] Log KV cache capacity after block-size resolution (#50462)
Tanmay Dixit committed 3 months ago
[Spec Decode] Add top-k DSpark Markov projection (#49969)
Andrii Skliar committed 3 months ago
[Docs] Fix two docs build warnings (#51014)
Harry Mellor committed 3 months ago
Support MLA properly in the Transformers modeling backend (#48250)
Harry Mellor committed 3 months ago
[XPU] fix collecting oneccl version info (#47104)
Yan Ma committed 3 months ago
[Bugfix] Resolve seq-cls `num_labels` from the top-level config for multimodal checkpoints (#50950)
Lazurite committed 3 months ago
[Bugfix][Build] Fix DeepGEMM CUDA 12.9 FP8 header visibility (#51003)
Kevin H. Luu committed 3 months ago
[Rust Frontend][gRPC] Add multimodal image inference (#50368)
Connor Carpenter committed 3 months ago
[Rust Frontend] Align tool rendering for Kimi K3 (#50540)
Bugen Zhao committed 3 months ago
[Rust][Benchmark] Preserve UTF-8 across benchmark stream chunks (#50868)
Reid committed 3 months ago
[XPU][CI]Adjust Samplers test ENV for Intel GPU (#50199)
xiangdong committed 3 months ago
[Bugfix][Reasoning] kimi_k3: O(delta) reasoning-end check on the decode path (#50886)
Bowen Wang committed 3 months ago
[ROCm][AITER] Hotfix for `memory access fault` errors in AITER triton MOE routing (#50859)
fxmarty-amd committed 3 months ago
[CI][Bugfix] Fix flaky `test_store_orders_after_compute_write` (#50926)
Nick Hill committed 3 months ago
Perf/h20 moe config e256 n512 (#48825)
zzt committed 3 months ago
[ROCm][Test] Use BF16 for Jina v5 nano MTEB test (#50917)
Andreas Karatzas committed 3 months ago
[Kimi-K3] Migrate FlashKDA to PyTorch stable ABI (#50818)
Thien Tran committed 3 months ago
[Bugfix][Kimi-K3] Enforce packed rows and op availability in AttnRes dispatch (#50567)
namgyu-youn committed 3 months ago
llmd+vllm+mori-ep(inter node wide-ep)+mori-io(write) for 2p2d with dp=ep=16 tp=1 (#45043)
shikamd123 committed 3 months ago
[Kernel] Extend CuTe DSL skinny GEMM to GLM-5.2 (#49791)
xiaozhoupy committed 3 months ago