[Model] Support DeepSeek-OCR-2 (#33165)

RED committed 9 months ago

Fix mistral sliding window parsing (#33521)

Andy Lo committed 9 months ago

[Doc]: update paths for Offline/Online/Others example sections (#33494)

Sawyer Bowerman committed 9 months ago

[Doc] add missing model entries in supported_models.md (#33220)

Paco Xu committed 9 months ago

[Bugfix] GLM-4 tool parser: incremental string streaming (#33218)

jack committed 9 months ago

[Nightly CI] Remove CT Model (#33530)

Robert Shaw committed 9 months ago

[Models] Step-3.5-Flash (#33523)

csy0225 committed 9 months ago

[Fix] prefix cache hit rate == 0 bug with gpt-oss style models (#33524)

Yifan Qiao committed 9 months ago

Add unpermute-aware fused MoE LoRA path (#32655)

Runkai Tao committed 9 months ago

[ModelRunner V2] Support spec decode with structured outputs (#33374)

Nick Hill committed 9 months ago

[ModelRunner V2] Misc minor simplifications and optimizations (#33467)

Nick Hill committed 9 months ago

[Misc] skip target model mm emb in draft proposal step when draft is text-only (#33437)

Komal Kumar Teru committed 9 months ago

Fix DeepSeek V2 RoPE initialization error (#33501)

will b. committed 9 months ago

Add MoE config for Super B200 TP2 (#33510)

shaharmor98 committed 9 months ago

[BUGFIX] Fix hipErrorIllegalState in Qwen3-Omni during startup profiling allow inference Omni on ROCM (#33077)

JartX committed 9 months ago

[W8A8 Block Linear Refactor][1/N] Keep all quantization types into `QuantFP8` class. (#33047)

Maral committed 9 months ago

[Redo] #33110 with threading limit (#33502)

Cyrus Leung committed 9 months ago

Change defaults for vllm bench startup (#33489)

Luka Govedič committed 9 months ago

fix: only include Authorization header when OPENAI_API_KEY is set (#33488)

Zack Yu committed 9 months ago

[Models]: lfm2_siglip2 return intermediate encoder layers (#33370)

Eduardo Salinas committed 9 months ago