{"slug": "jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding", "title": "Jina AI Releases jina-ocr-v1: A 3.4B MoE Document Parser With Built-In Speculative Decoding for Low-Budget GPUs", "summary": "Jina AI released jina-ocr-v1, a 3.4B-parameter Mixture-of-Experts visual document parser with roughly 570M active parameters per token, built on DeepSeek-OCR, that converts PDFs, scans, tables, charts and invoices into Markdown. The model includes a built-in FastMTP speculative decoding head that drafts 3 tokens per step while keeping output lossless, scoring 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR-Bench and parsing 2.57 pages per second on a single A100. Weights are available on Hugging Face under CC BY-NC 4.0, with hosted access through Jina Reader.", "body_md": "Jina AI has released jina-ocr-v1, a visual document parser that converts PDFs, scans, tables, charts and invoices into Markdown. The model has 3.4B total parameters, with about 570M active per token, and builds on DeepSeek-OCR. A built-in FastMTP speculative decoding head drafts 3 tokens per step while keeping output lossless. It scores 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR-Bench, and parses 2.57 pages per second on 1 A100. Weights are on Hugging Face under CC BY-NC 4.0, with hosted access through Jina Reader.\n\nThe post [Jina AI Releases jina-ocr-v1: A 3.4B MoE Document Parser With Built-In Speculative Decoding for Low-Budget GPUs](https://www.marktechpost.com/2026/09/18/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-speculative-decoding-for-low-budget-gpus/) appeared first on [MarkTechPost](https://www.marktechpost.com).", "url": "https://wpnews.pro/news/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding", "canonical_source": "https://www.marktechpost.com/2026/09/18/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-speculative-decoding-for-low-budget-gpus/", "published_at": "2026-09-18 21:21:59+00:00", "updated_at": "2026-09-18 21:25:27.166017+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research", "ai-products", "computer-vision", "ai-tools"], "entities": ["Jina AI", "jina-ocr-v1", "DeepSeek-OCR", "FastMTP", "OmniDocBench v1.6", "olmOCR-Bench", "Hugging Face", "Jina Reader"], "alternates": {"html": "https://wpnews.pro/news/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding", "markdown": "https://wpnews.pro/news/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding.md", "text": "https://wpnews.pro/news/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding.txt", "jsonld": "https://wpnews.pro/news/jina-ai-releases-jina-ocr-v1-a-3-4b-moe-document-parser-with-built-in-decoding.jsonld"}}