Perplexity open-sources next-gen decider models in 27B and 4B sizes with up to 5x faster inference
antoine_chaffin · x · 2026-10-17
Perplexity's antoinechaffin announces pplx-decider-v1.2 (27B) and pplx-decider-flash-v1.2 (4B), open-weight multimodal decision models. Text and vision scores improve; inference is up to 5x faster since all questions in a request run in one pass via a new attention schema. Flash is the top 4B on the Decision Index.
Related event: Perplexity Open-Sources pplx-decider-v1.2 in 27B and 4B Sizes(11 posts)→
More from Models
- Simulated robotics eval: Claude Opus 5.5 preferred over Opus 5 in 14 of 16 comparisons — qinzytech · 2026-10-17
- Unverified Watcher.Guru claim of Claude designing quantum circuits spreads as meme — jaivinwylde · 2026-10-17
- Google's Gemini 4 Argon 1M-Token Output Is Undersold for Legacy Modernization — rseroter · 2026-10-17
- Theo: Most DMs I Get Are 'Which $20 Plan Is Better, Claude or Codex?' — jxnlco · 2026-10-17
- Epoch AI: open-weight cyber capabilities trail closed frontier by just 6-7 months as GLM-5.3 outscores GPT-5.4 — teortaxesTex · 2026-10-17
- Span-02 sweeps chess engine Jev 12-0 with reasoning off, community asked to verify — victormustar · 2026-10-17