d1-omni-600M
d1-omni-600M is a 587M experimental decision model.
Research release
What the release provides
It combines a bidirectional text backbone with vision and audio encoders. It accepts text with images or up to 30 seconds of speech. Answers come from option probabilities, without generated text. The LFM Open License applies.d1-omni-600M official model card
How to use it well
Practical guidance. Use this research checkpoint for bounded decisions that fit its input limits. Inspect cases where the correct choice is absent from the supplied options.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | LiquidAI/d1-omni-600M |