Perplexity Developers

Official @perplexity_ai developer account for real-time API updates.

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.2 is the highest scoring decision model on ClaimBench.

Try pplx-decider-v1.2 and pplx-decider-flash-v1.2 on OpenRouter or Perplexity API Console.

https://t.co/6HlP9lsdpP

[Vals AI @ValsAI]

Today we’re releasing ClaimBench, our new benchmark targeted towards decision models. It tests whether a model can tell if a financial document supports, contradicts, or doesn’t address a claim.

We evaluated 15 models on 1,960 claims paired with 10-K excerpts, with Decider V1.2 leading the pack of decision models.

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.2 (27B) and pplx-decider-flash-v1.2 (4B) are now available on OpenRouter.

[OpenRouter @OpenRouter]

Perplexity Decider v1.2 and Decider Flash v1.2 are now on OpenRouter

@perplexity_ai's open-weights decision models take text or images and return typed answers with probabilities

- Decider v1.2: a new 27B checkpoint, succeeding v1.1
- Decider Flash v1.2: 4B, faster and half the cost

https://t.co/0Z7Do8gbjE
https://t.co/CknQd6rZLx

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.2 ranks #2 overall on the Hugging Face Decision Index.

Open weights for both models are available to run on your own infrastructure.

https://t.co/eI4FBcBvxg

https://t.co/u6dyifHu95

[Perplexity Developers @perplexitydevs]

Our next generation of open weights multimodal decision models comes in two sizes, pplx-decider-v1.2 and pplx-decider-flash-v1.2.

Text and vision scores improve. Inference is up to 5x faster: all questions in a request run in one pass.

Flash is the top 4B on the Decision Index.

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.2 is the 27B default for nuanced decisions at $0.02 per million input tokens.

pplx-decider-flash-v1.2 is for high-volume decisions with clear criteria like routing, labeling, and filtering, at half the price.

https://t.co/yS3ExNznmg

[Perplexity Developers @perplexitydevs]

Perplexity is now available in Lovable to give users real-time answers from the web, with citations.

[Lovable @Lovable]

If your app doesn’t have AI features you’re ngmi

We made them simple to build: just ask Lovable.

Now powered by even more models from @claudeai, @firecrawl, @GeminiApp, @OpenAI, @perplexity_ai, @p0, and @typesafeai.

[Perplexity Developers @perplexitydevs]

The Decisions API accepts text and image inputs.

Our new cookbook builds a browser agent that sends a screenshot and questions to pplx-decider-v1.1-27b. Your code turns the returned probabilities into clicks, scrolls, and stops.

https://t.co/yl6whABLzJ

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.1-27b scores the highest on Decision Bench for accuracy, with the lowest cost.

It reached 94.5% accuracy across 1071 cases at $0.017 per 1,000 decisions.

Full evaluation: https://t.co/uaJNEd6xaP
Get Started: https://t.co/f6RtdvJVHR

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.1-27b is now available on OpenRouter.

[OpenRouter @OpenRouter]

Perplexity Decider v1.1 is now on OpenRouter

@perplexity_ai's open-weights multimodal decision model takes text, JSON, or images and returns typed answers with probabilities

$0.02/M input. Output is free

https://t.co/PTD9b4aJZv

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.1-27b has a 250k context window and supports text and image input.

View the weights on Hugging Face:
https://t.co/qhx35AdiBL

Get started with the Perplexity Decisions API:
https://t.co/xdkKrOB6b2

[Perplexity Developers @perplexitydevs]

pplx-decider-v1.1-27b, our updated open weights multimodal decision model, is now available.

It scores the highest on the new @huggingface Decision Index 0.3 benchmark. pplx-decider-v1.1-27b costs half as much as v1, at $0.02 per million input tokens.

https://t.co/S6mjFL7Ws8

[Perplexity Developers @perplexitydevs]

We're looking forward to working with the new @NVIDIA Vera CPU on SPACE.

[Paul Fryzel @paulfryzel]

It was great to have the @NVIDIA team stop by and drop off their new Vera CPU.

A lot of what we’re building depends on agents being able to run more capable code environments, safely and reliably. We’re excited to test this work on Vera.

[Perplexity Developers @perplexitydevs]

You can use pplx-decider-v1-27b to triage customer support tickets, then route escalations to the Agent API.

Our cookbook shows how to build the workflow, using the Decisions API for the initial decision and the Agent API for escalated requests.

https://t.co/LVRwF5AviK

[Perplexity Developers @perplexitydevs]

Introducing the Perplexity Decisions API.

It's powered by pplx-decider-v1-27b, our multimodal decision model trained to output a probability distribution over a fixed set of answers instead of text.

It costs $0.04/million input tokens and scores 85.71% across benchmarks.

[Perplexity Developers @perplexitydevs]

pplx-decider-v1-27b is fine-tuned from Qwen3.8-27B with a 250k context window.

View the weights on Hugging Face:
https://t.co/miWYtmmXBZ

Get started with the Perplexity Decisions API:
https://t.co/kEWloEVpH4

[Perplexity Developers @perplexitydevs]

We created a cookbook recipe to test Profiles, Skills, and managed connectors together.

Follow it to build an incident responder that reads Datadog telemetry, diffs code in Sandbox, and posts an update to Slack.

https://t.co/I9mXa537at

[Perplexity Developers @perplexitydevs]

You can now build custom reusable agents in the Perplexity Agent API with Profiles, Skills, and managed connectors.

Configure an agent once in the API Portal and reuse it across applications and workflows.

[Perplexity Developers @perplexitydevs]

A Profile saves the agent's model, instructions, tools, Skills, and managed connectors under one versioned ID.

Project members can reuse the same tools and instructions without rebuilding the configuration in every application.

Read more about Profiles:
https://t.co/4nuLA1kBhu

[Perplexity Developers @perplexitydevs]

Fast Search is now the default search in Hermes Agent for Nous Portal subscribers.

It’s built for agentic tasks, with single-search-call latency of 160 ms at p50 and 230 ms at p95.

Learn more: https://t.co/irbFl40353