Perplexity Releases Decider V1.3 Models with Text-Span Extraction

Perplexity launched Decider v1.3 and Decider Flash v1.3 models featuring verbatim text-span extraction and downloadable Apache-2.0 weights, achieving 91.6% accuracy on 63 query-classification tasks. The models support precise text extraction for tasks like named-entity recognition and are accessible via OpenRouter and Perplexity’s Decisions API.