[CPU] Bump up CPU kernels to latest version (#50387)

Li, Jiang committed 3 months ago

[New model] Kimi K3 (#50000)

Jiangyun Zhu committed 3 months ago

[Test][ROCm] Account for gfx950 FP8 RMSNorm rounding (#49839)

Andreas Karatzas committed 3 months ago

[CI] Stabilize speculator memory teardown (#50284)

Andreas Karatzas committed 3 months ago

[CI][ROCm] Stabilize LLM GC teardown check (#50340)

Andreas Karatzas committed 3 months ago

[CI] Improve comment-triggered authorization and retries (#50414)

Kevin H. Luu committed 3 months ago

[Frontend] Lazily initialize chat media connectors (#49914)

Andreas Karatzas committed 3 months ago

[CI] Retry failed steps on new PR commits (#50318)

Kevin H. Luu committed 3 months ago

[DOC][CPU] remove tcmalloc warning from CPU docs (#50308)

Fadi Arafeh committed 3 months ago

[ROCm] Add AITER FP8 ViT encoder attention (#49937)

yinfengLiu committed 3 months ago

[XPU][CI]Add back skipped V1 test (#50207)

xiangdong committed 3 months ago

Add FlashMLA H100 tests to CI, fix them after #32810 (#50322)

Jane (Yuan) Xu committed 3 months ago

[ROCm] [CI] Support cached K/V (key/value=None) in Triton prefix-prefill (#48257)

stefankoncarevic committed 3 months ago

[CI/Build] Limit wheel size check to CUDA 13 (#50357)

Tyler Michael Smith committed 3 months ago

[Quantization][Autoround][XPU] Add W4A16(moe) / MXFP4(linear/moe) Support (#47124)

lkk committed 3 months ago

[PD][Bugfix] Rebase KV lease deadlines onto worker clock (#50326)

Nick Hill committed 3 months ago

[BugFix] Fix P/D preemption race condition (#50297)

Nick Hill committed 3 months ago

[Frontend] Add detokenization streaming derender for disaggregated serving (#47301)

Martin Hickey committed 3 months ago

[Bugfix][Kernel] Fix integer overflow in libtorch_stable/activation_kernels.cu (#49660)

Molly committed 3 months ago

Fix: FusedMoE AssertionError with Speculative Decoding on Quark-Quantized Models (#38293)

Venkateswara Rao Cherukuri committed 3 months ago