[Bugfix] Raise `VLLMValidationError` from structured output validators (#52394)

Jeffrey Wang committed 2 months ago

[Bugfix] Pick the DeepSeek V4 eager cudagraph region per model runner (#52401)

Nick Hill committed 2 months ago

[Bugfix][Spec Decode] Keep EAGLE cache registration on the partial-hash-hit path (#52419)

mispa-ms committed 2 months ago

[Bugfix][DSv4] Revert adaptive C128A metadata packing (#51318)

Toby Mao committed 2 months ago

[EC Connector] Added Build Connector Worker Meta for EC Connector (#49585)

omerpaz95 committed 2 months ago

[Bugfix][Sampling] Clear empty side on thinking-budget asymmetric SWAP (#49613)

Henry Su committed 2 months ago

[Bugfix][Model] Kimi-K3 MegaMoE: pass situ_beta/situ_linear_beta to fp8_fp4_mega_moe (#52445)

Uranus committed 2 months ago

[CI/Build] Add warning for unsupported global PTX architecture requests in... (#51901)

Shane Widanagama committed 2 months ago

[Bugfix] Make DSV4 sparse MLA work end-to-end for plain decode, MTP, and DSpark (#51538)

Gabriel Wu committed 2 months ago

[Frontend] Support count_reasoning_tokens in the Streaming Parser Engine (#45802)

Chauncey committed 2 months ago

[Bugfix] Fix modelscope usage (#52431)

Cyrus Leung committed 2 months ago

[Rust Frontend][gRPC] Preserve skip_special_tokens decoding option (#52384)

Biswa Panda committed 2 months ago

[ROCm][CI] Select CPU platform for native no-GPU jobs (#49515)

Andreas Karatzas committed 2 months ago

[Bugfix][Spec Decode] DSpark: inherit the target's attention backend when the speculative config names none (#52288)

Yongye Zhu committed 2 months ago

[Core] Check for GPU<->CPU syncs during CI (#43107)

Nick Hill committed 2 months ago

[Spec Decode][Perf] Fuse the MTP trailing all-reduce; local-argmax draft tokens (#49793)

xiaozhoupy committed 2 months ago

[MRV2] Support attention-free models (#52374)

Nick Hill committed 2 months ago

[ROCm]: Drop pybind11 from Dockerfile.rocm to prevent version mismatch (#52400)

Rohan Potdar committed 2 months ago

[CI] Shard Humming H100 eval (#52326)

Kevin H. Luu committed 2 months ago

[CI] Shard MoE refactor B200 eval (#52327)

Kevin H. Luu committed 2 months ago