[Bugfix][TPU][V1] Fix recompilation (#15553)

Nicolò Lucchesi committed over 1 year ago

Use absolute placement for Ask AI button (#15628)

Harry Mellor committed over 1 year ago

[Misc] Avoid direct access of global `mm_registry` in `compute_encoder_budget` (#15621)

Cyrus Leung committed over 1 year ago

[Feature] Add middleware to log API Server responses (#15593)

Yuan Tang committed over 1 year ago

[Misc] Replace `is_encoder_decoder_inputs` with `split_enc_dec_inputs` (#15620)

Cyrus Leung committed over 1 year ago

[Doc] Link to onboarding tasks (#15629)

Cyrus Leung committed over 1 year ago

[Bugfix] Fix use_cascade_attention handling for Alibi-based models on vllm/v1 (#15211)

Hiroaki Sugiyama committed over 1 year ago

[Model] MiniCPM-V/O supports V1 (#15487)

Cyrus Leung committed over 1 year ago

[Doc] update --system for transformers installation in docker doc (#15616)

Reid committed over 1 year ago

Fix incorrect filenames in vllm_compile_cache.py (#15494)

Richard Zou committed over 1 year ago

[Misc] Use model_redirect to redirect the model name to a local folder. (#14116)

wang.yuqi committed over 1 year ago

[Misc] Clean up `scatter_patch_features` (#15559)

Cyrus Leung committed over 1 year ago

[Quantization] Fp8 Channelwise Dynamic Per Token GroupedGEMM (#15587)

Robert Shaw committed over 1 year ago

[Misc] Consolidate LRUCache implementations (#15481)

Bella kira committed over 1 year ago

[TPU] Avoid Triton Import (#15589)

Robert Shaw committed over 1 year ago

[Misc] Restrict ray version dependency and update PP feature warning in V1 (#15556)

Rui Qiao committed over 1 year ago

[TPU] [V1] fix cases when max_num_reqs is set smaller than MIN_NUM_SEQS (#15583)

Chengji Yao committed over 1 year ago

[ROCm] Env variable to trigger custom PA (#15557)

Gregory Shtrasberg committed over 1 year ago

Allow torchao quantization in SiglipMLP (#15575)

Jerry Zhang committed over 1 year ago

[V1] Refactor num_computed_tokens logic (#15307)

Cody Yu committed over 1 year ago