Classification & Routing
Intent detection, ticket routing, content moderation, and triage.
Scoring & Rating
Urgency scoring, quality assessment, LLM-as-judge replacement.
Binary Gates
Safety checks, guardrails, extraction validation, yes/no decisions.
Multimodal
Visual inspection, image classification, voice-command routing.
d1 Models Latest release
d1-3B
3.1B · RecommendedMultimodal decision model (text + images) built on LFM2.5-VL-3B.
d1-omni-600M
587M · SmallestMultimodal decision model (text + images or audio) built on LFM2.5-Encoder-350M.
d1
Hosted · APIHosted decision model with Python and TypeScript SDKs via the Liquid API.
Primitives
Every question you ask a decision model uses one of three types:- Noul: A yes/no question that returns a probability between 0 and 1. “Is this message spam?” might return
0.92. - Choice: A pick-one-from-many question that returns a probability distribution over named options.
- Score: A rate-on-a-scale question that returns a probability-weighted position on an ordered rubric.
When to Use Decision Models
Decision models are a good fit when the answer is a structured decision: classification, categorization, routing (tickets, emails, requests), scoring, triage, content moderation, guardrails and safety checks, LLM-as-judge replacement, agent tool-call approval, and model routing or cascades. With image input, the same primitives support visual inspection, image classification, and decisions based on screenshots. With audio (d1-omni-600M), they support voice-command routing and spoken-intent detection. Use a language model instead for free-form text generation, creative writing, multi-turn conversation, complex multi-step reasoning, open-ended Q&A, summarization, and code generation. To move existing LLM calls to a decision model, see the Decision Model Guide.Examples
Build a Game with Decision Models
Pixel-art racing game using d1 Choice for lane selection.Platform: Desktop
