Perplexity Context API is live.
Context API reads up to 10 URLs per call and returns clean .md, HTML, links, and snippets per page.
It returns usable content for 92.3% of URLs, 87.9% fact recall, and 80.3% content precision at $1/1000 URLs fetched.
pplx.ai/context-api
Yesterday, I revealed that Perplexity crossed $1B ARR. We have strong margins and a healthy growth rate. We don’t talk about our finances very often because we’re focused on building. Follow our research: research.perplexity.ai
Today we're launching Perplexity Frontier Research.
The goal of Perplexity Frontier Research is to create a system that can perform autonomous research.
The work will focus on multi-agent orchestration, reinforcement learning, continual learning, systems, and pre-training.
We will develop algorithms, models, and evaluations, released openly, and tested in Perplexity products. Our work will be supported by more than 10,000 GPUs.
The Frontier Research team will regularly publish papers, models, data, and benchmarks.
We're actively hiring researchers and engineers for Perplexity Frontier Research in San Francisco, Palo Alto, New York, and London.
research.perplexity.ai
Today we're launching Perplexity Frontier Research.
The goal of Perplexity Frontier Research is to create a system that can perform autonomous research.
The work will focus on multi-agent orchestration, reinforcement learning, continual learning, systems, and pre-training.
We will develop algorithms, models, and evaluations, released openly, and tested in Perplexity products. Our work will be supported by more than 10,000 GPUs.
The Frontier Research team will regularly publish papers, models, data, and benchmarks.
We're actively hiring researchers and engineers for Perplexity Frontier Research in San Francisco, Palo Alto, New York, and London.
research.perplexity.ai
Teammates are now available in the Perplexity Mac app for Max subscribers.
Teammates take on work for you. They work individually or with other teammates to handle QA, customer service, research, growth marketing, and more.
one of the capabilities we are pushing very hard on for the decider model is security. we've already found many internal applications for it, and we now have a strong data flywheel that lets us improve in this direction very rapidly.
Neat example of how Perplexity used Decider internally
In one day it read 1B tokens across 7,000 agent sessions and found flaws that led to 14 fixes to their agent security rules.
Both models are Apache-licensed and free on OpenRouter until Nov 19
We use Decider internally to monitor our RL rollouts.
We consistently see better and more reliable results than using smaller API models like Luna or Haiku, and it's *way* cheaper.
We're rolling out pplx-decider-v1.3 and pplx-decider-flash-v1.3, the latest version of our open-weight decision models.
v1.3 adds locate, a new feature that extracts spans based on type or instructions.
Now available in our Decisions API, @huggingface, and @OpenRouter.
perplexity.ai/hub/blog/perpl…
training NER models on LLM labels also teaches them the LLM’s mistakes. Our error-type-aware loss masks incorrect labels during training, improving F1 by up to 4.6 points on Wikigold without extra training or clean validation labels.
We’re releasing v1.3 of our decider models today with improved performance and a new locate feature enabling you to extract information from your input. Check it out, it’s free on OpenRouter for the next week.
We're rolling out pplx-decider-v1.3 and pplx-decider-flash-v1.3, the latest version of our open-weight decision models.
v1.3 adds locate, a new feature that extracts spans based on type or instructions.
Now available in our Decisions API, @huggingface, and @OpenRouter.
perplexity.ai/hub/blog/perpl…
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.
We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view.
pplx-embed-v2-context-9b-preview sets a new state of the art on ConTEB and @turbopuffer's new, privately held context-bench.
perplexity.ai/hub/blog/conte…
I'll be in Hamburg this week at KONVENS where Pascal Stolzenburg will present our work on news-crawler-LM (arxiv.org/abs/2607.21284)!
At our chair, we develop Fundus (github.com/flairnlp/fundus), a fast and high-quality news crawler using tailored rules for each publisher.