d1-omni-600M

Use it for tasks such as routing, triage, moderation, classification, extraction checks, reranking, agent guardrails, and visual inspection.

Availability

Model available on Hugging Face with loading instructions

Pricing model

Not established

Deployment

587M parameters; described as edge-sized

Privacy

Not established

Licence

other

• The model returns typed answers directly from its distribution over options, with zero output tokens. ^[1]

  answers with **zero output tokens**: every answer is read directly from the model's distribution over the

• It supports text and images in a single forward pass. ^[1]

  - **Vision-language**: text and images (tiled for large frames, several images per state) in a single forward pass.

• It supports text and up to 30 seconds of speech in a single forward pass. ^[1]

  - **Audio-language**: text and up to 30 s of speech in a single forward pass.

• The model can be loaded from LiquidAI/d1-omni-600M using Transformers with trust_remote_code=True. ^[1]

  The model ships its own code, so load it with `trust_remote_code=True`:

• The source says the model is not a chat model and does not write text. ^[1]

  It is not a chat model and does not write text.

• Liquid AI’s decision models classify inputs without writing a response Daily brief 2026-10-07