Perplexity Launches Open-Weight Decider v1.3 Models Featuring New Locate Feature
Perplexity has introduced Decider v1.3 and Flash v1.3, open-weight decision models that output probabilities rather than generated text, alongside a new Locate feature for extracting precise text spans.
Key points
- Perplexity released pplx-decider-v1.3 and flash-v1.3 as open-weight models.
- The models output probabilities and scores rather than generating text.
- A new Locate feature extracts exact text spans for tasks like PII detection.
Search and AI startup Perplexity has rolled out its latest open-weight decision models, designated as pplx-decider-v1.3 and its smaller counterpart, pplx-decider-flash-v1.3. Unlike standard large language models designed to generate conversational text, these specialized systems evaluate text and image inputs to return precise probabilities for yes/no, multiple-choice, and scoring questions.
The primary architectural upgrade in version 1.3 is the introduction of Locate, a tool built to pinpoint exact text spans using short labels or custom instructions, making it suited for tasks such as named entity recognition and personally identifiable information detection. The standard model runs on a 27-billion-parameter backbone, while the Flash variant operates on a 4-billion-parameter base, with model weights released under the Apache 2.0 license on Hugging Face.
According to Perplexity's internal evaluations, the full-size model achieved a mean accuracy of 91.6% across 63 query-classification tasks. The models are also accessible via the Perplexity Decisions API and OpenRouter, requiring hardware capable of accommodating roughly 49 GiB of BF16 weights for the primary version and 8.5 GiB for the Flash edition.
What to watch next: Monitor third-party benchmarking and developer adoption to see how these open-weight probability models perform outside of Perplexity's self-reported metrics.
Source MarkTechPost — marktechpost.com/2026/11/13/perplexity-decider-v1-3-open-weight-dec… ↗
Read next
🧠AI AnalysisSalesforce Optimizes Global AI Model Inference with Amazon SageMaker
Salesforce has implemented cross-Region inference routing in Amazon SageMaker to manage AI model capacity across 30+ global regions more efficiently.
🏢Big TechMicrosoft Challenges Rivals on AI Pricing at Ignite Conference
Microsoft is positioning its Copilot AI as the more cost-effective choice for businesses compared to OpenAI and Anthropic, citing a commissioned study to support its clai…
🚀Startups & FundingOpenAI CEO Sam Altman Addresses Upcoming Biographical Film
OpenAI CEO Sam Altman commented on the upcoming film 'Artificial,' which portrays his life and the 2023 leadership crisis at the company.
🔌Chips & HardwareSnapchat Integrates AI Coding Assistants into AR Lens Development
Snapchat has launched the Easy Lens CLI, enabling developers to build augmented reality experiences using AI coding tools like Claude Code and Cursor.
🚀Startups & FundingRadixArk Unveils DaisyRose-27B, Focused on Agentic Efficiency
Startup RadixArk has launched DaisyRose-27B, an AI model optimized for long-horizon agent tasks, claiming significant improvements in training efficiency and reasoning ca…
🔌Chips & HardwareUS Eyes the Philippines for Massive AI and Chip Hub Amid Domestic Opposition
President Donald Trump visited the Philippines to advance the Pax Silica initiative, proposing a 1,620-hectare hub for data centers and chip fabs to bypass local pushback…