Perplexity introduces pplx-decider-v1.2 and pplx-decider-flash-v1.2
Perplexity says text and vision scores improved, with inference up to 5x faster.
TLDR
Perplexity announced two open-weight multimodal decision models: pplx-decider-v1.2 and the smaller pplx-decider-flash-v1.2. It says all questions in a request run in one pass, inference is up to 5x faster, and Flash is the top 4B model on the Decision Index. A developer says the models add reranking and are available in the API, on OpenRouter and as open source; new “latest” slugs point to the most recent versions.
Combined views
19.6K
7 Sources, first seen ago
