AIMarkTechPostabout 3 hours ago
vLLM Releases Decision 3.0 Models
1 min read
The vLLM Semantic Router team released Decision 3.0, a family of five multimodal decision models ranging from 0.85B to 26.09B parameters. These Apache-2.0 licensed models read text, JSON and images to answer typed questions. They run on 1 AMD Instinct MI325X GPU using BF16 weights to classify and route requests without parsing generated text.
Level
Hype check
Tap to vote and see what everyone thinks.
