[CustomOp] Support object-level enable for CustomOp (#30547)

Shanshan Shen committed 11 months ago

[Log] Skip piecewise cudagraph warn when using full cudagraph (#30657)

Boyuan Feng committed 11 months ago

Revert "[Fix]Load kv-cache dtype from hf_quant_config.json automatically" (#30653)

Robert Shaw committed 11 months ago

CPU KV Offloading: Use more CUDA streams (#29013)

Or Ozeri committed 11 months ago

[responsesAPI]add extra body parameters (#30532)

RioS committed 11 months ago

[NIXL][BUG FIX] Fix a bug for PD with host_buffer after merging 29665 (#30420)

Chendi.Xue committed 11 months ago

[Model][Quantization] Override HF defaults to GGUF ones (incl. Qwen3 MoE) (#30118)

Tsukasa OI committed 11 months ago

[Bugfix][benchmarks] Fix input token calculation for rerank benchmark metrics (#30596)

Vensen committed 11 months ago

[Feature]Add EVS (Efficient Video Sampling) Support for Qwen3-VL (#29752)

ZiTian Zhao committed 11 months ago

[Bugfix] Improve error messages in ModelConfig validation (#30213)

yifant-code committed 11 months ago

[Bugfix] Fix fusion for VL models (#30244)

ElizaWszola committed 11 months ago

[Bugfix] Revert Qwen2-VL part of change in #28271 (#30542)

zifeitong committed 11 months ago

[Bugfix] Update get_processor_data to use get_all method (#30583)

Drew Botwinick committed 11 months ago

[Misc] Add a script to benchmark compilation time (#29919)

Bin Bao committed 11 months ago

[Bugfix] fix _get_quant_method of FusedMoE for deepseekV3.2 on non-NV… (#30057)

tjp_zju committed 11 months ago

Nvidia ModelOpt workaround for issue 28072 (#30164)

Shengliang Xu committed 11 months ago

fix: Update json features supported by xGrammar (#30390)

Johannes F committed 11 months ago

[Bugfix] awq_gemm: fix argument order swap (#30364)

Matthias Gehre committed 11 months ago

[torch.compile] Add encoder tag for compilation (#30489)

Ilya Markov committed 11 months ago

Add AudioFlamingo3 model support (#30539)

Lasha Koroshinadze committed 11 months ago