{"slug": "paddleocr-vl", "name": "PaddleOCR-VL", "category": "other", "category_name": "Other software", "description": "", "recommenders": 3, "commenters": 5, "warners": 0, "score": 3.376, "rising": 1.67, "first": "2026-01-20", "latest": "2026-08-21", "why": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 2, "compare": 0, "warn": 0}, "spark": [0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html", "kind": "", "icon": true, "domain": null, "github_repo": null, "official_url": "https://github.com/PaddlePaddle/PaddleOCR", "pricing": "", "aliases": ["PaddleOCR-VL"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 790194672, "item_id": 49387167, "url": "https://news.ycombinator.com/item?id=49387167", "by": "dandaka", "date": "2026-08-21", "stance": "recommend", "quote": "I use mlx-community/PaddleOCR-VL-8bit", "excerpt": "but for OCR there are much better suited models, I use mlx-community/PaddleOCR-VL-8bit", "context": "", "use_case": "OCR of paper documents", "switched_from": "", "story": "DeepSeek-v4-flash-vision-exp"}, {"id": 789991542, "item_id": 49374471, "url": "https://news.ycombinator.com/item?id=49374471", "by": "cpnwaugha", "date": "2026-08-20", "stance": "recommend", "quote": "you can switch between DeepSeek OCR 2, GLM-OCR, dots.mocr, Paddle OCR VL, PP-OCRv6, etc.", "excerpt": "Yeah, GPT 5.6 Sol is very good. Generally, Gemini models remain SOTA for VLM tasks with 3.7-flash at the top. --- That said, considering variables like cost (say, over 100k PDF pages) and accuracy requirements (e.g., construction documents with dense images & tables), Gemini and other SOTA VLMs are expensive and inaccurate, and therefore unsuitable. This is where niche, open-weight, and task-specific OCR/VL models co", "context": "", "use_case": "", "switched_from": "", "story": "GPT 5.6 Sol is the best \"vision\" model OpenAI ever released"}, {"id": 785069408, "item_id": 49066838, "url": "https://news.ycombinator.com/item?id=49066838", "by": "TimurKramar", "date": "2026-07-27", "stance": "recommend", "quote": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html", "excerpt": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html#3-model-performance", "context": "", "use_case": "OCR", "switched_from": "", "story": "Show HN: Open-weight OCR got so cheap I had to share it"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["ocr of paper documents", "ocr"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49386163", "title": "DeepSeek-v4-flash-vision-exp", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49329575", "title": "GPT 5.6 Sol is the best \"vision\" model OpenAI ever released", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49041317", "title": "Show HN: Open-weight OCR got so cheap I had to share it", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48344012", "title": "Show HN: CPU-only fast OCR for screenshots, images, PDFs, webpages", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46691454", "title": "Show HN: Ocrbase \u2013 pdf \u2192 .md/.json document OCR and structured extraction API", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"neutral": 1}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {"neutral": 1}, "2026-07": {"recommend": 1}, "2026-08": {"recommend": 2}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 790194672, "item_id": 49387167, "url": "https://news.ycombinator.com/item?id=49387167", "by": "dandaka", "date": "2026-08-21", "stance": "recommend", "quote": "I use mlx-community/PaddleOCR-VL-8bit", "excerpt": "but for OCR there are much better suited models, I use mlx-community/PaddleOCR-VL-8bit", "context": "", "use_case": "OCR of paper documents", "switched_from": "", "story": "DeepSeek-v4-flash-vision-exp"}, {"id": 789991542, "item_id": 49374471, "url": "https://news.ycombinator.com/item?id=49374471", "by": "cpnwaugha", "date": "2026-08-20", "stance": "recommend", "quote": "you can switch between DeepSeek OCR 2, GLM-OCR, dots.mocr, Paddle OCR VL, PP-OCRv6, etc.", "excerpt": "Yeah, GPT 5.6 Sol is very good. Generally, Gemini models remain SOTA for VLM tasks with 3.7-flash at the top. --- That said, considering variables like cost (say, over 100k PDF pages) and accuracy requirements (e.g., construction documents with dense images & tables), Gemini and other SOTA VLMs are expensive and inaccurate, and therefore unsuitable. This is where niche, open-weight, and task-specific OCR/VL models co", "context": "", "use_case": "", "switched_from": "", "story": "GPT 5.6 Sol is the best \"vision\" model OpenAI ever released"}, {"id": 785069408, "item_id": 49066838, "url": "https://news.ycombinator.com/item?id=49066838", "by": "TimurKramar", "date": "2026-07-27", "stance": "recommend", "quote": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html", "excerpt": "World class: https://www.paddleocr.ai/main/en/version3.x/algorithm/PaddleOCR-VL/PaddleOCR-VL-1.6.html#3-model-performance", "context": "", "use_case": "OCR", "switched_from": "", "story": "Show HN: Open-weight OCR got so cheap I had to share it"}, {"id": 773712576, "item_id": 48357036, "url": "https://news.ycombinator.com/item?id=48357036", "by": "mrkn1", "date": "2026-06-01", "stance": "neutral", "quote": "It should support 109 languages", "excerpt": "It should support 109 languages. More info here: https://huggingface.co/PaddlePaddle/PaddleOCR-VL", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: CPU-only fast OCR for screenshots, images, PDFs, webpages"}, {"id": 747143840, "item_id": 46696490, "url": "https://news.ycombinator.com/item?id=46696490", "by": "cess11", "date": "2026-01-20", "stance": "neutral", "quote": "The OCR model looks kind of small, https://huggingface.co/PaddlePaddle/PaddleOCR-VL/tree/main", "excerpt": "Why is 12GB+ VRAM a requirement? The OCR model looks kind of small, https://huggingface.co/PaddlePaddle/PaddleOCR-VL/tree/main, so I'm assuming it is some processing afterwards it would be used for.", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Ocrbase \u2013 pdf \u2192 .md/.json document OCR and structured extraction API"}]}