Migrate LlavaNextVideoPixelInputs to TensorSchema (#21843)

Benji Beck committed over 1 year ago

enable Docker-aware precompiled wheel setup (#22106)

Doug Smith committed over 1 year ago

Fix TensorSchema validation test for symbolic dims (#22366)

Benji Beck committed over 1 year ago

Remove redundant row_indices unsqueeze operation in MiniCPMO (#22528)

ZiTian Zhao committed over 1 year ago

Migrate LlavaNextImageInputs to TensorSchema (#21774)

Benji Beck committed over 1 year ago

Fix(benchmarks): allow multiple mm contents in OpenAI Chat Completion Benchmarks (#22534)

Breno Baldas Skuk committed over 1 year ago

[Bugfix][Kernel] Support partial rotary embedding for MRoPE triton kernel (#22593)

Isotr0py committed over 1 year ago

[doc] add alibaba cloud as sponsor (#22597)

youkaichao committed over 1 year ago

[doc] add beijing meetup links (#22596)

youkaichao committed over 1 year ago

Move `CacheConfig` from `config/__init__.py` to `config/cache.py` (#22586)

Harry Mellor committed over 1 year ago

[Misc] Replace flaky image urls in pixtral test (#22574)

Isotr0py committed over 1 year ago

[Docs] Fix warnings in docs build (#22588)

Harry Mellor committed over 1 year ago

[Misc] Further refine type annotations in parallel state (#22499)

Cyrus Leung committed over 1 year ago

[Doc] Fix API doc link in side navigation (#22585)

22quinn committed over 1 year ago

[Misc] code clean duplicate set_current_vllm_config in _set_vllm_config (#22566)

Ning Xie committed over 1 year ago

[Minor] Fix pre-commit error on main (#22579)

Isotr0py committed over 1 year ago

Refactor sliding window configuration to Transformers best practice (#21927)

Harry Mellor committed over 1 year ago

[TPU] kv cache update kernel doesn't need to be padded slices to multiple of num_slices_per_block (#22394)

Chengji Yao committed over 1 year ago

Improve fast_topk function with type hints and documentation (#22530)

ZiTian Zhao committed over 1 year ago

[Config] add "qwen" as a native eagle3 target supported model (#22333)

Le Chen committed over 1 year ago