Add Triton recompile detection (#45631)

Thien Tran committed 5 months ago

[Misc]Clean up useless test (#45792)

wangxiyuan committed 5 months ago

[XPU][CI] add intel xpu cases for nightly CI (#44372)

wenjun liu committed 5 months ago

[Bugfix] Fix trtllm fused allreduce+rms_norm for transformers backend (#45307)

Thomas Parnell committed 5 months ago

[Frontend] Remove AsyncMicrobatchTokenizer. (#45759)

wang.yuqi committed 5 months ago

[Bugfix] Prevent cuMemcpyBatchAsync segfault with MTP and KV offloading (#44784)

joshua abraham committed 5 months ago

[CPU] Support Gemma Diffusion (#45690)

Li, Jiang committed 5 months ago

[Bug Fix] Allow pinned memory for WSL2 (#41496)

Jimmy Lee committed 5 months ago

[Cleanup] Remove dead env (#45777)

Cyrus Leung committed 5 months ago

[Misc] Added validation for Cohere /v2/embed input field exclusivity (#45640)

Taneem Ibrahim committed 5 months ago

Register parsed config classes before tokenizer init (#40299)

Andrew Barnes committed 5 months ago

[Core] Use fastsafetensors ParallelLoader for weight loading (#40183)

gitbisector committed 5 months ago

[Frontend] [Parser] Migrate Nemotron V3 to streaming parser engine (#45755)

Ben Browning committed 5 months ago

[Rust Frontend] Require `ModelConfig.vocab_size` to be present (#45696)

Bugen Zhao committed 5 months ago

[Bugfix][Gemma4] Fix offline parser truncation, adjust_request token leak, and chat template sync (#45553)

Luciano Martins committed 5 months ago

[XPU] Fix Triton attn fp8/bf16 check failing (#45758)

liuzhenwei committed 5 months ago

[BugFix] Support async scheduling with prompt embeds for multimodal models (#45673)

Ruinan Ma committed 5 months ago

[Multimodal] Add Qwen3-VL video loader (#44412)

Isotr0py committed 5 months ago

[Rust Frontend] Lower out-of-vocab validation to `text` layer (#45685)

Bugen Zhao committed 5 months ago

[Rust Frontend] Support `max_logprobs` validation (#45674)

Bugen Zhao committed 5 months ago