Use the same memory for workspace13 and fused_output. (#31531)

Andrey Khalyavin committed 10 months ago

[CI] Move Distributed Tests from H200 -> H100 (#32555)

Robert Shaw committed 10 months ago

[MoE Refactor] Separate Router into OO Classes (#30623)

bnellnm committed 10 months ago

"refactor: refactor_repeated_interfaces" (#32486)

tjp_zju committed 10 months ago

[Bugfix] Fix GLM-ASR audio encoder RoPE dim (#32540)

Isotr0py committed 10 months ago

[Model] Support Step1 Model (#32511)

Li Xie committed 10 months ago

[Model] Remove the unnecessary dtype conversion in MiniCPM (#32523)

Canlin Guo committed 10 months ago

[Model Runner V2] Minor optimization for eagle input processing (#32535)

Woosuk Kwon committed 10 months ago

[Performance] Improve Triton prefill attention kernel's performance (#32403)

Isotr0py committed 10 months ago

[MoE Refactor] Move Test Impl into Test Dirs (#32129)

Robert Shaw committed 10 months ago

[Model Runner V2] Move mrope_positions buffer to MRopeState (#32532)

Woosuk Kwon committed 10 months ago

[Feature] Add FIPS 140-3 compliant hash algorithm option for multimodal hashing (#32386)

Karan Bansal committed 10 months ago

[build] fix cu130 related release pipeline steps and publish as nightly image (#32522)

Shengqi Chen committed 10 months ago

[CI/Build][Docker] Add centralized version manifest for Docker builds (#31492)

Mritunjay Kumar Sharma committed 10 months ago

[Misc] Fix typo: seperator -> separator in flashmla_sparse.py (#32411)

Guofang.Tang committed 10 months ago

[Model] Molmo2: Enable quantized weight mapping for vision backbone (#32385)

Kim Hee Su committed 10 months ago

[Models] Lfm2Moe: minor name changes for resolving lora conflicts (#29063)

Paul Pak committed 10 months ago

[CI] Implement uploading to PyPI and GitHub in the release pipeline, enable release image building for CUDA 13.0 (#31032)

Shengqi Chen committed 10 months ago

Revert "[Attention][MLA] Make `FLASHINFER_MLA` the default MLA backen… (#32484)

Matthew Bonanni committed 10 months ago

apply _validate_input to MistralTokenizer token-id chat prompts (#32448)

vanshil shah committed 10 months ago