Perplexity's Decider v1.3 lets one model classify, score, and extract in a single pass
Perplexity Decider v1.3: Training, eval, and use cases
Perplexity released Decider v1.3 and a distilled Flash v1.3, decision models that return probability distributions over yes/no, multiple-choice, or ordered-score questions in one forward pass. Built on Qwen backbones with bidirectional attention and a three-stage training recipe, they add a locate feature that extracts named entities and PII as character spans. Both ship under Apache license on Hugging Face, OpenRouter, and the Decisions API.
Decision models sit between LLMs and task-specific classifiers. They return probabilities for multiple questions about the same input in a single forward pass, avoiding token-by-token answer generation and sharing computation across questions.
- 0o777
This blog post acted as a rabbit hole into some interesting stuff:
https://www.perplexity.ai/hub/blog/pii-trace-detecting-perso...
https://www.perplexity.ai/hub/blog/a-local-first-agent-for-p...
- waximabbax
Well congrats.
Here are some thoughts not facts:
On AA Decision Index.
Decider v1.3 and Decider v.1.2 seem to have same benchmarks. However, V1.3 flash beats V1.2 flash by 4 points.
I am not particulary excited about this pattern/trend of having flash versions for decision models. The whole point of Decision models was to deliver faster ouput for high volume requests at much cheaper cost than Reasoning models. Since most reasoning models can and probably do outperform decision models already in accuracy but they are a lot slower and are pricier. Reasoning Models can already return structured ouput.
So your flagship model must be fastest and most accurate.
Unless it can handle some different form of input other than just image and text, I don't see the point of having both flagship and flash version for the same model.
- nchmy
Because of this post, i discovered and dug into various decider and other models. The "locate" mechanism in Decider 1.3 seems to have an equivalent in Clairic from cloudflare and Extract from ScaleDown. I'm particularly impressed with scaledown for my use case - much cheaper and more accurate than the others.