Liquid AI · Decision
d1.2
d1.2 is Liquid AI's default decision model: instead of generating text it reads text and images and returns calibrated, typed answers in one forward pass with no output tokens — yes/no, one or several choices, a score, a value (text, number or date) or every passage of a kind read from the input, a box around one object or every instance in a picture, or a number estimated with a range. At less than half d1.2-ultra's price, for classification, routing, scoring, extraction, redaction, visual inspection and LLM-as-judge checks.
Advantages and trade-offs
Compared with 10 representative decision models
Trade-offs
Input & output
Inputs as stated by providers; output from the model’s typeText
Images
Choices and scores
Also known as: liquid/d1.2
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works