AI BriefingPerplexity Decider V1.2 models now available on AI Gateway
Key point
The new multimodal decision models, Decider V1.2 27B and Decider Flash V1.2 4B, classify content and route requests via typed questions.
Details
Perplexity has released Decider V1.2 27B and Decider Flash V1.2 4B on AI Gateway. These multimodal decision models are designed for classifying content, routing requests, and scoring text, structured records, and images.
Model Capabilities and Variants
The models answer typed questions with probabilities, supporting Boolean, Choice, and Score question types. Applications can use these outputs to route support tickets, flag damaged products, or escalate uncertain decisions for human review. Multiple questions can share the same input in a single request.
- Decider V1.2 27B (
perplexity/pplx-decider-v1.2): The larger model, optimized for nuanced judgments and detailed rubrics. - Decider Flash V1.2 4B (
perplexity/pplx-decider-flash-v1.2): The smaller model, designed for high-volume labeling, filtering, and routing with clear criteria.
Integration and Pricing
Developers can integrate these models using the AI SDK, the Decisions API, or the TypeSafe-compatible API. For image inputs, base64 image_url objects can be embedded in JSON state. AI Gateway allows users to call models with a single API key, monitor usage, and track costs. Inference uses provider pricing without markup or platform fees, even when using BYOK (Bring Your Own Key) credentials.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.