Move to a faster base64 implementation (#19984)

h-avsha committed over 1 year ago

Revert "Fix(models/siglip): Add compatibility for Gemma models quantized by llm-compressor" (#20030)

Isotr0py committed over 1 year ago

Revert "[Feature] Integrate new deepgemm (#19820)" (#20049)

Wentao Ye committed over 1 year ago

[Llama4] Update `attn_temperature_tuning` (#19997)

Brayden Zhong committed over 1 year ago

cmake: Update vllm_flash_attn for vllm_kernels (#20032)

Eli Uriegas committed over 1 year ago

use .dev for version comparison with pytorch nightly release (#20031)

Boyuan Feng committed over 1 year ago

[Easy] Remove submodule added in #19463 (#20039)

Brayden Zhong committed over 1 year ago

[Feature] Integrate new deepgemm (#19820)

Wentao Ye committed over 1 year ago

[P/D] Asynchronously do _nixl_handshake (#19836)

lkchen committed over 1 year ago

[BugFix] Fix multi-node offline data parallel (#19937)

Nick Hill committed over 1 year ago

[Misc][Benchmarking] Add variable request-rate ("ramp-up") to the benchmarking client. (#19423)

d.transposed committed over 1 year ago

[Perf] Improve/Fix-regression for FA3 in High QPS regimes (#19463)

Lucas Wilkinson committed over 1 year ago

[Fix][V1] Remove --scheduling-policy oracle (#20010)

amit committed over 1 year ago

refactor example - qwen3_reranker (#19847)

Reid committed over 1 year ago

[Bugfix][CPU] Fix InputBatch for pooling models in the CPU v1 (#20014)

Li, Jiang committed over 1 year ago

[PERF] Speedup of MRoPE prepare inputs (#19939)

Vadim Gimpelson committed over 1 year ago

add some examples for other benchmark scripts (#19893)

Reid committed over 1 year ago

[doc] Fix broken link in the installation for CPU (#19980)

Kay Yan committed over 1 year ago

[Misc][Tools][Benchmark] Add profile to autotune script (#19711)

Chenyaaang committed over 1 year ago

Update test case parameter to have the throughput above 8.0 (#19994)

QiliangCui committed over 1 year ago