vLLM's Semantic Router team has released d3-ultra, a 27B open multimodal decision model under Apache 2.0. It writes no text: it reads text, images or video once and returns a probability for every answer, so one call covers routing, classification and visual grounding. It scored 87.2 on the team's SR-Text test, ahead of GLiDE V1.2 on 85.1. https://www.marktechpost.com/2026/10/26/vllm-semantic-router-d3-ultra-27b-multimodal-decision-model/ #AIagent #AI #GenAI #AIresearch
Related
Hiltswals (2026) by Jamie Bradbury (J Bizzel) for ZX Spectrum. Proper scoring, seeded games you can share with friends a...
Hiltswals (2026) by Jamie Bradbury (J Bizzel) for ZX Spectrum. Proper scoring, seeded games you can share with friends and a generous undo. Press ENTER to change the felt colour wh...
Claude and Codex can help us build products faster than ever. Can AI handle marketing as well as it handles coding? Yes ...
Claude and Codex can help us build products faster than ever. Can AI handle marketing as well as it handles coding? Yes or No?#AI #AIMarketing #BuildInPublic
China’s trade surplus passed $1 trillion in 10 months as AI demand lifted exportsChina's trade surplus hit $1.04 trillio...
China’s trade surplus passed $1 trillion in 10 months as AI demand lifted exportsChina's trade surplus hit $1.04 trillion in the first 10 months, a month earlier than last year, wi...