Liquid D1 Ultra
Liquid AI's largest decision model, served as a System One endpoint: it reads text and images and answers typed questions with a choice, a score or a yes/no probability.
From Liquid, D1 Ultra is a cloud decision model. It reads a piece of text and answers typed questions about it, such as a choice between labels, a yes or no, or a score on a scale, with a probability for every option instead of written text. It reads up to 33K input tokens per request. It runs through OpenRouter using your own API key, from $0.042 per million input tokens.
- Released
- Oct 2026
- Pricing
- $0.042 / 1M in · $0.00 / 1M out
- Answers
- Choice, yes/no, score
- Context window
- 33K in
Samples
Answered by D1 Ultra via OpenRouter. A decision model returns probabilities, not prose, so each card shows how sure it was about every option. The same four questions run against every Decision model in the catalog.
Hi, I was charged twice for my March subscription. Can you refund the duplicate payment before the end of the week?
Which team should handle this support ticket?
- Payments, invoices and refunds100%
- Errors or broken features in the product0%
- Login, passwords and account settings0%
- Plans, pricing questions and upgrades0%
海辺のカフェの写実的な写真、夕暮れ、35mmフィルム風
What kind of image does this prompt ask for? It can be in any language.
- A logo or brand mark0%
- A photorealistic shot99%
- Drawn, painted or stylised art0%
- Anything else1%
What is a vector database, and when would a small team need one? I'm not asking you to build anything yet.
Does this message ask the assistant to create or generate something?
- Yes: it asks for something to be made0%
- No: it asks a question or wants an explanation100%
Checkout has returned an error for every customer since 9am and we are losing orders. Please look now.
How urgent is this message?
- Can wait for the next release0%
- Should be handled this week0%
- Should be handled today1%
- Needs attention right now99%
Liquid
Liquid develops Liquid Foundation Models (LFM) — a new architecture optimized for efficient long-context reasoning.
www.liquid.ai ↗Liquid AI's mixture-of-experts reasoning model with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools.
Liquid's structured decision model, served as a System One endpoint for choice, score and yes/no answers.
Liquid's 24B foundation model with efficient long-context reasoning.
Liquid's efficient 8B Liquid Foundation Model tuned for long-context reasoning on-device.
Liquid AI's 600M decision model on LFM2.5-Encoder-350M, running fully offline: it answers typed questions about a text or JSON state with a choice, a score or a yes/no probability.