--- format: "aidr-story-markdown/v1" id: "4910a1bd9135f12c130aecc0648b9e268c8e5cdbbfa5f87cf024ffad8514b7b0" canonical_url: "https://aidr.today/4910a1bd?lang=en" title: "Perplexity AI Releases pplx-embed-v2-late: A 0.6B Edge Model and a 9B Model Scoring 92.4% on MADQA" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-08T05:39:09.000Z" category: "Data" topics: ["llm","data-engineering","microsoft","open-source"] source_urls: ["https://www.marktechpost.com/2026/10/07/perplexity-ai-releases-pplx-embed-v2-late-a-0-6b-edge-model-and-a-9b-model-scoring-92-4-on-madqa/","https://www.marktechpost.com/2026/10/09/microsoft-ai-releases-microsoft-decision-1-a-qwen3-5-9b-decision-scoring-model/","https://www.marktechpost.com/2026/10/09/nace-ai-open-sources-drex-1-5-a-9b-decision-model-that-scores-options-not-text/"] summary: "Perplexity AI has released pplx-embed-v2-late, which includes a 0.6B model for edge devices and a 9B model for building indexes. Both MIT-licensed models scored 92.4% on MADQA and 61.2% on ViDoRe v3 Markdown." --- # Perplexity AI Releases pplx\-embed\-v2\-late: A 0\.6B Edge Model and a 9B Model Scoring 92\.4% on MADQA > [Open the canonical story]() **Published:** 2026-10-08T05:39:09.000Z **Category:** Data **Topics:** llm, data\-engineering, microsoft, open\-source ## Summary Perplexity AI has released pplx\-embed\-v2\-late, which includes a 0\.6B model for edge devices and a 9B model for building indexes\. Both MIT\-licensed models scored 92\.4% on MADQA and 61\.2% on ViDoRe v3 Markdown\. ## Sources - [Story source]() - [Story source]() - [Story source]()