[Bugfix] Fix platform-specific routing in CustomOp implementations (#24444)

Konrad Zawora committed about 1 year ago

[Kernels] Enable Torch Symmetric Memory All-Reduce By Default (#24111)

Ilya Markov committed about 1 year ago

[VLM] Optimize GLM4.5-V-style video processing to only decode necessary frames (#24161)

Isotr0py committed about 1 year ago

[Docs] Fix typos in EP deployment doc (#24669)

Harry Mellor committed about 1 year ago

[Doc] Fix Markdown Pre-commit Error (#24670)

Wentao Ye committed about 1 year ago

[Docs] Add transcription support to model (#24664)

Nicolò Lucchesi committed about 1 year ago

[Bugifx] Fix qwen-next packed_modules_mapping (#24656)

Jee Jee Li committed about 1 year ago

[HybridKVCache][Platform] Add support_hybrid_kv_cache for platform (#24646)

Mengqing Cao committed about 1 year ago

Allow users to specify kv cache memory size (#21489)

Boyuan Feng committed about 1 year ago

[CI] Split mteb test from Language Models Test (#24634)

wang.yuqi committed about 1 year ago

[Bench] Add qwen-next in benchmark_moe.py (#24661)

Jee Jee Li committed about 1 year ago

[build] add torch to tool.uv no-build-isolation-package (#24303)

youkaichao committed about 1 year ago

[Docs] Fixes a typo in the qwen3next model name. (#24654)

Tao He committed about 1 year ago

[XPU] add missing dependency tblib for XPU CI (#24639)

Fanli Lin committed about 1 year ago

Move `LoRAConfig` from `config/__init__.py` to `config/lora.py` (#24644)

Harry Mellor committed about 1 year ago

Fix typing for `safetensors_load_strategy` (#24641)

Harry Mellor committed about 1 year ago

[CI Failure] fix models/language/pooling/test_auto_prefix_cache_support.py (#24636)

wang.yuqi committed about 1 year ago

[Misc] Add @NickLucche to codeowners (#24647)

Nicolò Lucchesi committed about 1 year ago

[CI] Split pooling from entrypoints Test (#24632)

wang.yuqi committed about 1 year ago

[Bugfix] Fix _synced_weight_loader (#24565)

Kyuyeun Kim committed about 1 year ago