← Back to live feed1 story
1
Liquid AI Releases Open-Weight d1-3B and d1-omni-600M: Multimodal Decision Models With Zero Output Tokens
marktechpost.comLiquid AI released two open-weight multimodal models in its d1 decision model family: d1-3B reads text and images, and d1-omni-600M reads text with an image or text with audio, with neither model writing text and each returning calibrated, typed answers in one forward pass with zero output tokens.
Sign in to suggest edits
Key sources
- SOURCE@openrouter“takes text, JSON, or images and returns typed answers with probabilities”x.com
- SOURCE@perplexitydevs“scores the highest on the new @huggingface Decision Index 0.3 benchmark”x.com
- SUPPORT@maziyarpanahi“pplx-decider v1.1 got 643 of 669 right vs 628 for Jev, at 42% less per decision”x.com
- SUPPORT@hwchase17“cheap typed answers for the small calls inside a harness (routing, approvals, judging)”x.com
- SUPPORT@denisyarats“trained on top of qwen3.8-27b, so it natively supports 250k context and is multimodal”x.com
- SUPPORT@alokbishoyi97“8 decision models playing Tetris. In our tests @perplexity_ai Decider came consistently at the top”x.com
- SOURCEhuggingnewshuggingnews.com
- SOURCEmarketbrief.now