vllm-sr releases Decision 3.0 multimodal decision models in six sizes from 0.6B to 26B
victormustar · x · 2026-10-12
vllm-sr released the Decision 3.0 collection on Hugging Face, positioned as open multimodal foundation decision models with zero-shot classification. The series spans six sizes: d3 (26B), d3-flash (8B), d3-mini (5B), d3-nano (2B), d3-lite (0.9B), and d3-edge (0.6B), all updated and downloadable. The collection also includes earlier Decision 1.0/2.0, Vela 1.0/2.0, and MoM series models.
Related event: vLLM Community Open-Sources Decision 3.0 Decision Model Family(5 posts)→
More from Models
- METR's AI time-horizon chart hasn't added a model since May 8 — mhmazur · 2026-10-12
- Claude Opus 5.5 vs GPT-6.1 Sol hands-on: this time GPT wins the vibe-check — CodeByPoonam · 2026-10-12
- DeepMind's AlphaProof Nexus reportedly cracks 9 Erdős problems, 44 OEIS conjectures — kimmonismus · 2026-10-12
- Codex keeps stalling on legit debugging over cybersecurity guardrails, user complains — Al_Grigor · 2026-10-12
- Third-party provider adds vision to Kimi K3, enables DeepSeek max thinking mode — sarahdrinkwater · 2026-10-12
- Google's inline AI outputs bizarre typo 'outshitting' for 'outshining' — felix_cohen · 2026-10-12