{"slug": "mistral-ocr", "name": "Mistral OCR", "category": "ai", "category_name": "AI models and assistants", "description": "", "recommenders": 4, "commenters": 8, "warners": 2, "score": 1.677, "rising": 1.35, "first": "2026-01-02", "latest": "2026-09-08", "why": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 3, "compare": 1, "warn": 2}, "spark": [0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 2, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly.", "icon": false, "domain": null, "github_repo": null, "official_url": "", "pricing": "", "aliases": ["Mistral OCR", "Mistral OCR 4"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 792622976, "item_id": 49538936, "url": "https://news.ycombinator.com/item?id=49538936", "by": "icantevenhold", "date": "2026-09-02", "stance": "recommend", "quote": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly.", "excerpt": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly. Of course I\u2019m not always on the most bleeding edge forefront of the latest hyped model so there might be better alternatives, but I\u2019m happy with what they provide", "context": "", "use_case": "OCR", "switched_from": "", "story": "Can I opt out of my input or output data being used for training?"}, {"id": 792587376, "item_id": 49536711, "url": "https://news.ycombinator.com/item?id=49536711", "by": "perks_12", "date": "2026-09-02", "stance": "recommend", "quote": "The OCR stuff is quite usable.", "excerpt": "The OCR stuff is quite usable. Their models perform good at some very basic tasks, so I use them to diversify. But their current model lineup is really terrible.", "context": "", "use_case": "OCR", "switched_from": "", "story": "Can I opt out of my input or output data being used for training?"}, {"id": 779859984, "item_id": 48741249, "url": "https://news.ycombinator.com/item?id=48741249", "by": "refulgentis", "date": "2026-06-30", "stance": "recommend", "quote": "OCR is off the charts good on every metric you can think of.", "excerpt": "OCR is off the charts good on every metric you can think of. LLMs are a near-afterthought at this point if you don\u2019t have data residency requirements. I love them and they\u2019re slightly underrated, their models are consistently well-trained, open, but as you note, behind. There is no metric that will say they\u2019re ahead in anything.", "context": "", "use_case": "OCR", "switched_from": "", "story": "Leanstral 1.5"}, {"id": 779856784, "item_id": 48741049, "url": "https://news.ycombinator.com/item?id=48741049", "by": "hakunin", "date": "2026-06-30", "stance": "recommend", "quote": "I use it because it\u2019s a simple, convenient and cheap OCR api.", "excerpt": "I use it because it\u2019s a simple, convenient and cheap OCR api. Specifically via my ringbinder[1] tool. [1]: https://github.com/maxim/ringbinder", "context": "", "use_case": "OCR API", "switched_from": "", "story": "Leanstral 1.5"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 793765616, "item_id": 49610351, "url": "https://news.ycombinator.com/item?id=49610351", "by": "hakunin", "date": "2026-09-08", "stance": "warn", "quote": "Mistral doesn't even come close on anything non-trivial.", "excerpt": "I've made a personal doc tool based on Mistral OCR API at first, and then switched to Gemini Flash for the same price inline, or half price using batch API, and the difference is night and day. Mistral doesn't even come close on anything non-trivial. Longer story here: https://max.engineer/ringbinder", "context": "built a personal doc tool", "use_case": "document OCR", "switched_from": "", "story": "Mistral raises \u20ac3B"}, {"id": 776696209, "item_id": 48543513, "url": "https://news.ycombinator.com/item?id=48543513", "by": "SyneRyder", "date": "2026-06-15", "stance": "warn", "quote": "I use regular Claude Sonnet for OCR because it is more accurate than Mistral OCR.", "excerpt": "I've tried using Mistral for various tasks, and it is so far behind the American models that I just never bother using it despite still having lots of Mistral API credits leftover. Even their OCR and TTS products are surpassed by generic US models - I use regular Claude Sonnet for OCR because it is more accurate than Mistral OCR. I could rant about this, I am just so disappointed at how Mistral completely gave up and", "context": "", "use_case": "OCR", "switched_from": "", "story": "Show HN: Can Europe train a frontier AI model on the compute it owns?"}]}], "who": [], "tags": [], "use_cases": ["ocr", "document ocr", "ocr api", "ocr of scanned paperback"], "compared": [{"slug": "gemini-flash", "name": "Gemini Flash", "co": 1, "switched_from": 0, "switched_to": 1}, {"slug": "claude-sonnet", "name": "Claude Sonnet", "co": 1, "switched_from": 0, "switched_to": 1}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49535284", "title": "Can I opt out of my input or output data being used for training?", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48738938", "title": "Leanstral 1.5", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48645152", "title": "Mistral OCR 4", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49605767", "title": "Mistral raises \u20ac3B", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48541014", "title": "Show HN: Can Europe train a frontier AI model on the compute it owns?", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48449187", "title": "Ask HN: What are tools you have made for yourself since the advent of AI", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"neutral": 1}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {"neutral": 2, "warn": 1, "compare": 1, "recommend": 1}, "2026-07": {}, "2026-08": {}, "2026-09": {"recommend": 2, "warn": 1}, "2026-10": {}}, "quotes": [{"id": 793765616, "item_id": 49610351, "url": "https://news.ycombinator.com/item?id=49610351", "by": "hakunin", "date": "2026-09-08", "stance": "warn", "quote": "Mistral doesn't even come close on anything non-trivial.", "excerpt": "I've made a personal doc tool based on Mistral OCR API at first, and then switched to Gemini Flash for the same price inline, or half price using batch API, and the difference is night and day. Mistral doesn't even come close on anything non-trivial. Longer story here: https://max.engineer/ringbinder", "context": "built a personal doc tool", "use_case": "document OCR", "switched_from": "", "story": "Mistral raises \u20ac3B"}, {"id": 792622976, "item_id": 49538936, "url": "https://news.ycombinator.com/item?id=49538936", "by": "icantevenhold", "date": "2026-09-02", "stance": "recommend", "quote": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly.", "excerpt": "Mistral OCR is really good and their small models are great for simple translations, I use them regularly. Of course I\u2019m not always on the most bleeding edge forefront of the latest hyped model so there might be better alternatives, but I\u2019m happy with what they provide", "context": "", "use_case": "OCR", "switched_from": "", "story": "Can I opt out of my input or output data being used for training?"}, {"id": 792587376, "item_id": 49536711, "url": "https://news.ycombinator.com/item?id=49536711", "by": "perks_12", "date": "2026-09-02", "stance": "recommend", "quote": "The OCR stuff is quite usable.", "excerpt": "The OCR stuff is quite usable. Their models perform good at some very basic tasks, so I use them to diversify. But their current model lineup is really terrible.", "context": "", "use_case": "OCR", "switched_from": "", "story": "Can I opt out of my input or output data being used for training?"}, {"id": 779859984, "item_id": 48741249, "url": "https://news.ycombinator.com/item?id=48741249", "by": "refulgentis", "date": "2026-06-30", "stance": "recommend", "quote": "OCR is off the charts good on every metric you can think of.", "excerpt": "OCR is off the charts good on every metric you can think of. LLMs are a near-afterthought at this point if you don\u2019t have data residency requirements. I love them and they\u2019re slightly underrated, their models are consistently well-trained, open, but as you note, behind. There is no metric that will say they\u2019re ahead in anything.", "context": "", "use_case": "OCR", "switched_from": "", "story": "Leanstral 1.5"}, {"id": 779856784, "item_id": 48741049, "url": "https://news.ycombinator.com/item?id=48741049", "by": "hakunin", "date": "2026-06-30", "stance": "recommend", "quote": "I use it because it\u2019s a simple, convenient and cheap OCR api.", "excerpt": "I use it because it\u2019s a simple, convenient and cheap OCR api. Specifically via my ringbinder[1] tool. [1]: https://github.com/maxim/ringbinder", "context": "", "use_case": "OCR API", "switched_from": "", "story": "Leanstral 1.5"}, {"id": 778341218, "item_id": 48646326, "url": "https://news.ycombinator.com/item?id=48646326", "by": "philipkglass", "date": "2026-06-23", "stance": "neutral", "quote": "If you have an OCR problem, Mistral OCR 4 is probably great.", "excerpt": "I used Abbyy Finereader for several years. I loved it. I completed some large projects with it. Modern VLMs put classic FineReader to shame for processing low-resolution/degraded/non-standard text. I'm personally using the small Qwen 3.5 models. If you have an OCR problem, Mistral OCR 4 is probably great. Open weights models that you can run on a laptop may also work great.", "context": "", "use_case": "OCR", "switched_from": "", "story": "Mistral OCR 4"}, {"id": 778333537, "item_id": 48645846, "url": "https://news.ycombinator.com/item?id=48645846", "by": "mdrzn", "date": "2026-06-23", "stance": "compare", "quote": "It'll be interesting to see how this ranks against", "excerpt": "It'll be interesting to see how this ranks against https://github.com/baidu/Unlimited-OCR", "context": "", "use_case": "OCR", "switched_from": "", "story": "Mistral OCR 4"}, {"id": 776696209, "item_id": 48543513, "url": "https://news.ycombinator.com/item?id=48543513", "by": "SyneRyder", "date": "2026-06-15", "stance": "warn", "quote": "I use regular Claude Sonnet for OCR because it is more accurate than Mistral OCR.", "excerpt": "I've tried using Mistral for various tasks, and it is so far behind the American models that I just never bother using it despite still having lots of Mistral API credits leftover. Even their OCR and TTS products are surpassed by generic US models - I use regular Claude Sonnet for OCR because it is more accurate than Mistral OCR. I could rant about this, I am just so disappointed at how Mistral completely gave up and", "context": "", "use_case": "OCR", "switched_from": "", "story": "Show HN: Can Europe train a frontier AI model on the compute it owns?"}, {"id": 775291873, "item_id": 48455742, "url": "https://news.ycombinator.com/item?id=48455742", "by": "hakunin", "date": "2026-06-08", "stance": "neutral", "quote": "as provided by Mistral OCR", "excerpt": "A SQLite based sweeper of all the scans, notes, PDFs and images I have on my filesystem, that stores their paths and allows searching their OCR\u2019ed descriptions and text, as provided by Mistral OCR. I can ask things like \u201cwhen does my car need maintenance\u201d or \u201cfind me that picture my kid drew for Mother\u2019s Day\u201d. I use pi-based bash executable to launch a doc chat like that. https://github.com/maxim/ringbinder", "context": "", "use_case": "OCR", "switched_from": "", "story": "Ask HN: What are tools you have made for yourself since the advent of AI"}, {"id": 743435552, "item_id": 46464722, "url": "https://news.ycombinator.com/item?id=46464722", "by": "raudette", "date": "2026-01-02", "stance": "neutral", "quote": "Even with modern OCR (I used Mistral's here)", "excerpt": "Related: for fun over the holidays, I created an ePub of a paperback copy of \"I Brought The Ages Home\", by Charles T. Currelly, which went out of copyright in Canada in 2007 (copyright in Canada changed from 50 to 70 years after the death of the author in 2022, but this did not affect works that were already in the public domain). I couldn't find a ebook online, so I found an old paperback copy and created one: https", "context": "", "use_case": "OCR of scanned paperback", "switched_from": "", "story": "Standard Ebooks: Public Domain Day 2026 in Literature"}]}