Perplexity Decisions API v1.2 — pplx-decider-v1.2 and decider-flash-v1.2
Perplexity released pplx-decider-v1.2 (27B, Qwen3.8-27B backbone) and pplx-decider-flash-v1.2 (4B), decision models that return typed probabilities instead of text.
Why it matters
Joint text+image decisions (images up to 1.6 MP) in one call at $0.02/M input tokens with free output — Decision Index 0.3 up from 62.30 to 68.09 over v1.1, and a 4B sibling at $0.01/M for high-volume routing.
What you could build with it
A customer-support startup could route every incoming ticket with one API call — picking the right team, flagging urgency, and scoring sentiment as probabilities — and let application code auto-assign or escalate based on confidence thresholds instead of parsing chat output.
Does it hold up?
Too early to judge — the API launched Oct 17; Ollama added day-one MLX support, but no independent production reports exist beyond Perplexity's own benchmarks.
Built with Perplexity Decisions API v1.2
- Perplexity's Decider V1.2 models now available on AI Gatewayvercel · Vercel's AI Gateway lists Decider V1.2 27B and Decider Flash V1.2 4B with the same request shape across the AI SDK, the Decisions API, and a TypeSafe-compatible endpoint.
- Perplexity Sets Input Fees for Two Decision Models at $0.01 and $0.02econotimes · Breakdown of the v1.2 pricing ($0.02/M for the 27B, $0.01/M for the 4B, free output), the grouped-vs-individual question guidance, and the Decision Index 0.3 scores 68.09 vs 62.30 for v1.1.
- ollama/ollamagithub · 183,792 ★ · v0.40.7-rc0 adds Decider v1.2 serving support on the MLX backend, one day after the API launch.
Learn more
First spotted on article: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73StepFun Step-5-Preview — open BF16 weights releasedStepFun published the full BF16 weights for its flagship Step-5-Preview agentic model (600B-parameter sparse MoE, 27B…models · JEV 0.73NVIDIA Nemotron 3.5 Retriever collection: Embed 8B/1B, Rerank 8B, Parse 631MNVIDIA's Oct 21 collection drop for agentic retrieval: Nemotron-3.5-Embed in 8B BF16, 1B BF16 and 1B NVFP4…models · JEV 0.72Kimi K3.1 Pro: Moonshot's 2.8T open-weight flagship officially launchesMoonshot AI officially launched Kimi K3.1 Pro on Oct 23, a mixture-of-experts model with 2.8 trillion total / 104…models · JEV 0.72
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.