[multimodal] Make PyNvVideoCodec decoder concurrency configurable (#49753)

Brandon Pelfrey committed 3 months ago

Stabilize GPU memory teardown between ROCm CI tests (#49242)

Aarushi Jain committed 3 months ago

[CI] fix compile test | refactor VLLM_DISABLE_COMPILE_CACHE for tests (#49770)

Divakar Verma committed 3 months ago

[ROCm][CI] Force native compile caches onto local disk (#49763)

Aarushi Jain committed 3 months ago

[Bugfix][KV Connector][Mooncake] Keep TP-sharded Mamba state out of the KV-head dedup (#49499)

Yifan Qiao committed 3 months ago

[XPU] add warning for xpu graph limitations (#49419)

liuzhenwei committed 3 months ago

[BUGFIX] Fix log capture in KV test (#49655)

liuzhenwei committed 3 months ago

Revert "[Perf][GLM-5.2] Blackwell decode optimizations" (#49768)

Woosuk Kwon committed 3 months ago

[ROCm][CI] Wait for ROCm VRAM to settle between compiled and eager LL… (#49739)

Aarushi Jain committed 3 months ago

[ROCm][Docker] Drop MORI_GPU_ARCHS so MoRI autodetects the device arch (#49737)

Rohan Potdar committed 3 months ago

[ROCm][CI] Fix XPASS(strict) on mixed audio embeds test (#49733)

djramic committed 3 months ago

[KV Offload][CI] Fall back to buffered I/O without O_DIRECT; fix flaky api-server test (#49734)

Harry Mellor committed 3 months ago

[Model] Remove Plamo2 (#49729)

Harry Mellor committed 3 months ago

[CI] Stabilize memory-sensitive compile and structured output tests (#49749)

Jiangyun Zhu committed 3 months ago

[Bugfix] Support non-uniform page sizes in KVBlockZeroer (#49704)

Elvir Crnčević committed 3 months ago

[AMD][Bugfix][EPLB] Fix elastic EP scaling accuracy on ROCm (#47206)

Oxana Korzh committed 3 months ago

[WideEP] Update NCCL to 2.30.7 to enable DeepEPv2 in the vllm/vllm-openai image (#45321)

Tyler Michael Smith committed 3 months ago

[Bugfix] Register axk1 config to fix A.X-K1 init (#49727)

djramic committed 3 months ago

Fix GLM-4.1V video placeholder token ID handling. (#49484)

Aarushi Jain committed 3 months ago

[CI] Avoid unnecessary Hugging Face metadata requests (#49508)

Andreas Karatzas committed 3 months ago