Decision 3.0 open multimodal decision models released

The vLLM Semantic Router team released Decision 3.0 (d3), a family of decision models from 0.6B to 27B under Apache-2.0 that read text, images, and video. It ranks #1 on both Jev Decision Index boards.

• Cloudflare's Clef model removes humans from AI agent loop2026-10-02
• Vercel data: Jev is fastest-adopted AI model ever2026-10-16
• China proposes export controls on 23 AI technologies2026-10-16
• User Builds Three 4x R9700 Local AI Servers with 384GB Total VRAM2026-10-15
• Mapping post-Transformer architecture landscape in 20262026-10-15