[torch.compile] Reorganize vllm/compilation and tests/compile (0/N for vLLM IR) (#33731)

Luka Govedič committed 9 months ago

[CPU][BugFix] Fix loading of w8a8int models with bias (#33582)

Fadi Arafeh committed 9 months ago

Bump HF Hub client to get bug fix (#33984)

Harry Mellor committed 9 months ago

[PaddleOCR-VL] Add BC for transformers 5.0 config (#33976)

zhang-prog committed 9 months ago

Consolidate and fix forbidden import `pre-commit` checks (#33982)

Harry Mellor committed 9 months ago

support view_from_cpu_tensor on XPU (#33868)

Xinyu Chen committed 9 months ago

Fix `main` pre-commit (#33975)

Harry Mellor committed 9 months ago

[cpu][performance] CPU Paged Attention NEON BFMMLA BF16 Implementation (#32263)

Gassan Salama committed 9 months ago

Onboard voyage-4-nano (#33720)

chengchengpei committed 9 months ago

[XPU]Replace pip in docker.xpu with uv pip (#31112)

sihao_li committed 9 months ago

[XPU][4/N] add mxfp4 moe model support (#33679)

Kunshang Ji committed 9 months ago

[CPU] Add BF16 Kernel type for s390x (#33788)

R3hankhan committed 9 months ago

[Misc] Update code for encoder-decoder models (#33900)

Cyrus Leung committed 9 months ago

feat(frontend): early-fail tokenization guard for user requests (#31366)

Mingliang Li committed 9 months ago

fix(ROCm): Make flash_attn import optional in MLA attention (#33511)

Rabi Mishra committed 9 months ago

[Docs] Add reo analytics (#33957)

Simon Mo committed 9 months ago

[Perf] Disable clean_logits in deepgemm fp8_mqa_logits kernel (#33568)

Xin Yang committed 9 months ago

[Feature] OTEL tracing during loading (#31162)

emricksini-h committed 9 months ago

[Bugfix] Fix DeepSeek v3.2 tokenizer outputting None issue (#33832)

Wei Zhao committed 9 months ago

Adds padding and perf improvements to wvSplitK_fp8 (#33527)

Hashem Hashemi committed 9 months ago