vLLM Semantic Router team open-sources 'Decision 3.0,' multimodal decision models that also analyze images (report)

Verified (fact)

The vLLM Semantic Router team open-sourced 'Decision 3.0,' a family of multimodal decision models that analyze images directly, as well as text and JSON. AI Times said in its headline that Cloudflare also released open-source models, but the excerpt contains no details on Cloudflare's release.

Why it matters (interpretation)

Competition among decision models used for routing and classification is expanding beyond text to multimodal inputs.

Sources

• Press 의사결정 모델도 멀티모달로 진화… vLLM·클라우드플레어 오픈소스 모델 공개 AI Times (Korea)

More about vLLM

• 2026-10-07 P2 Open source Generally available vLLM
  vLLM v0.31.0 released, focused on DeepSeek-V4.1-Flash inference performance

  vLLM v0.31.0 includes 717 commits from 307 contributors (96 of them new). The main change is better DeepSeek-V4.1-Flash performance:...