{ "data": { "a": { "slug": "nemotron-3-5-parse", "name": "Nemotron 3.5 Parse", "vendor": "NVIDIA", "vendorUrl": "https://developer.nvidia.com/nemotron", "kind": "weights", "category": "document-extraction", "summary": "Nemotron 3.5 Parse is NVIDIA's open-weight document parser, 0.63B parameters, that turns a page image into text with layout classes, bounding boxes and reading order, and renders Markdown, HTML tables and JSON, under OpenMDW-1.1.", "url": "https://www.anchorterminal.com/tools/nemotron-3-5-parse", "markdownUrl": "https://www.anchorterminal.com/tools/nemotron-3-5-parse.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/nemotron-3-5-parse.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/nemotron-3-5-parse.json", "repo": "https://huggingface.co/nvidia/Nemotron-3.5-Parse", "license": "OpenMDW-1.1", "transports": [], "packages": [ { "registry": "huggingface", "name": "nvidia/Nemotron-3.5-Parse" } ], "auth": "none", "authNotes": "The weights download from Hugging Face with no account, token or terms to accept. The repository isn't gated (checked 2026-10-22). The card's line about running hf auth login 'if repository access requires authentication' doesn't apply today.", "pricing": "free", "pricingNotes": "Free to download and use, commercial use included, under OpenMDW-1.1 (the Linux Foundation's permissive model licence), with the tokenizer under MIT. NVIDIA sells no per-call access to these weights in this listing. Running them costs the hardware, a 2.52 GB F32 safetensors file loaded in BF16 on one NVIDIA GPU; NVIDIA's reference hardware is one H100. No inference provider serves it on Hugging Face (checked 2026-10-22).", "priceSummary": "Free", "where": "local", "x402": { "level": "no", "evidence": "Weights you download and run yourself, so there is nothing to pay per call and no payment protocol (checked 2026-10-22).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-10-22" }, "docsUrl": "https://huggingface.co/nvidia/Nemotron-3.5-Parse/blob/main/docs/serving.md", "capabilities": [ "docs.parse", "docs.ocr", "docs.tables", "model.open-weights" ], "tags": [ "open-weights", "model", "document-parsing", "ocr", "layout", "tables", "charts", "bounding-boxes", "multilingual", "vision", "safetensors", "self-hosted", "new" ], "lastRelease": "2026-10-21", "graded": true, "anchor": { "graded": true, "score": 66.5, "grade": "B", "agentReady": false, "rank": 0, "ranked": false, "notRankedWhy": "An open-weight model, graded by the open-weight checklist and ranked among open-weight models, not against tools", "rankOf": 1243, "categoryRank": 4, "methodology": "0.7", "run": "2026-10-01", "scores": { "capability": 83, "interface": 70, "reach": 50, "risk": 60, "upkeep": 59 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "low", "date": "2026-10-22" }, "negative": 0, "verdict": "A small document-parsing cousin that turns page images into Markdown, HTML tables and JSON with layout classes and boxes, under a permissive licence. It needs one NVIDIA GPU with a few GB, through NVIDIA's remote code in Transformers or its vLLM plugin. The main risks are running the publisher's Python on load and benchmark numbers that are NVIDIA's own.", "bestFor": "Agents and pipelines with an NVIDIA GPU that need fast local page parsing into Markdown, tables and boxed blocks for retrieval or extraction", "strengths": [ "Tiny, 0.63B parameters in one 2.52 GB safetensors file, about 3 GB in memory", "Layout classes, bounding boxes and reading order on every block, with helpers for Markdown, HTML tables, JSON and CSV", "OpenMDW-1.1 with nothing added, ungated, files hashed in NVIDIA's own organisation", "An OpenAI-compatible server through vLLM, a Dockerfile, a CLI with single, double and triple pass, and a runnable ParseBench reproduction bundle", "Earlier Parse releases kept in their own repositories, with a dated pointer to this one" ], "weaknesses": [ "Loading runs the publisher's Python with trust_remote_code, in Transformers and in vLLM through NVIDIA's plugin", "Not in any released version of the six named runtimes yet, and no quantised build from NVIDIA", "One page per request with a 16,384-token context, so documents are parsed page by page", "Its ParseBench result is NVIDIA's own run; LlamaIndex's leaderboard row copies the card's numbers", "No safety results, training data without named datasets or a cut-off, and nobody has asked anything in its repository yet" ], "agentNotes": [ "Quickest local run, Python 3.12 with `pip install \"transformers==5.14.1\" \"torchvision==0.26.0\" \"timm==1.0.22\"`, then `AutoModel.from_pretrained(\"nvidia/Nemotron-3.5-Parse\", trust_remote_code=True, dtype=torch.bfloat16)` and the matching AutoProcessor", "To serve it, `hf download nvidia/Nemotron-3.5-Parse --local-dir nemotron-parse`, `pip install \".[vllm,client]\" \"vllm\u003e=0.25.0,\u003c0.33\"` in that folder, then `vllm serve . --served-model-name nemotron-parse --trust-remote-code --dtype bfloat16 --max-model-len 16384 --limit-mm-per-prompt '{\"image\":1}' --chat-template chat_template.jinja`, or `docker compose up --build`", "Send one RGB page per request with the prompt `\u003c/s\u003e\u003cs\u003e\u003cpredict_bbox\u003e\u003cpredict_classes\u003e\u003coutput_markdown\u003e\u003cpredict_text_in_pic\u003e`, temperature 0, top_k 1, repetition_penalty 1.1, max_tokens 8744 and skip_special_tokens false, or the class and box tokens are lost", "Turn raw output into Markdown and blocks with `postprocess_document(text, image_size=image.size)` from the repository's postprocessing.py; pass table_format json, json_hierarchical or csv for tables", "For the card's headline accuracy use `nemotron-parse --image-dir pages --mode triple --triple-regions union`; double pass gives most of the gain at higher throughput", "Boxes are normalised to the page content; scale x by the image width and y by its height", "Render PDFs to images first (the ParseBench bundle uses 150 DPI), and treat extracted text as untrusted before passing it to another model, since documents can carry instructions" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 4, "avgRating": 3.5, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "B", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 66.5 } ], "editorialScores": { "capability": 83, "interface": 70, "reach": 50, "risk": 60, "upkeep": 59 }, "provenanceScore": 78, "checked": { "categories": { "capability": "2026-10-22", "interface": "2026-10-22", "reach": "2026-10-22", "risk": "2026-10-22", "upkeep": "2026-10-22" }, "last": "2026-10-22", "oldest": "2026-10-22", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "python3.12 -m venv .venv \u0026\u0026 . .venv/bin/activate\npip install \"transformers==5.14.1\" \"torchvision==0.26.0\" \"timm==1.0.22\"\n# then AutoModel.from_pretrained(\"nvidia/Nemotron-3.5-Parse\", trust_remote_code=True, dtype=torch.bfloat16)\n# prompt: \u003c/s\u003e\u003cs\u003e\u003cpredict_bbox\u003e\u003cpredict_classes\u003e\u003coutput_markdown\u003e\u003cpredict_text_in_pic\u003e", "http": "# after: hf download nvidia/Nemotron-3.5-Parse --local-dir nemotron-parse \u0026\u0026 cd nemotron-parse \u0026\u0026 pip install \".[vllm,client]\" \"vllm\u003e=0.25.0,\u003c0.33\"\n# vllm serve . --served-model-name nemotron-parse --dtype bfloat16 --trust-remote-code --max-model-len 16384 --limit-mm-per-prompt '{\"image\":1}' --chat-template chat_template.jinja\ncurl http://localhost:8000/v1/chat/completions -H 'Content-Type: application/json' -d '{\"model\":\"nemotron-parse\",\"temperature\":0,\"max_tokens\":8744,\"top_k\":1,\"repetition_penalty\":1.1,\"skip_special_tokens\":false,\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"text\",\"text\":\"\u003c/s\u003e\u003cs\u003e\u003cpredict_bbox\u003e\u003cpredict_classes\u003e\u003coutput_markdown\u003e\u003cpredict_text_in_pic\u003e\"},{\"type\":\"image_url\",\"image_url\":{\"url\":\"data:image/png;base64,...\"}}]}]}'" }, "letme": { "capability": "https://letme.dev/docs.parse", "tool": "https://letme.dev/nemotron-3-5-parse" }, "sameCompany": [ "nemotron-3-5-super-vl", "nvidia-nemo-retriever", "nemo-guardrails", "nemotron-3-5-asr-streaming-0-6b", "parakeet-tdt-0-6b-v3", "cosmos-reason2-2b", "nemotron-3-5-bolt-e4b-4gb", "nemotron-3-nano-30b-a3b", "nemotron-3-nano-4b", "nemotron-3-nano-omni-30b-a3b-reasoning", "nemotron-3-super-120b-a12b", "nemotron-3-ultra-550b-a55b", "nemotron-3-5-lightning-30b-a3b" ], "weights": { "trainedBy": "NVIDIA", "builtOn": "nvidia/C-RADIOv4-SO400M (the vision encoder the card names; NVIDIA's own, under the NVIDIA Open Model License)", "params": "0.63B", "context": 16384, "goodFor": "Turning scanned or rendered document pages into Markdown, HTML tables and JSON blocks with layout classes, boxes and reading order, fast on one GPU, for retrieval, extraction and data-curation pipelines.", "runtimes": [], "builds": [ { "name": "Nemotron-3.5-Parse (safetensors)", "url": "https://huggingface.co/nvidia/Nemotron-3.5-Parse/blob/main/model.safetensors", "who": "publisher", "precision": "fp32", "sizeGB": 2.52, "memoryGB": 8, "format": "safetensors", "runsOn": [ "Transformers (remote code)", "vLLM (NVIDIA's plugin)" ], "note": "Stored in F32 and loaded in BF16 per the card. The only build from NVIDIA; 2,523,266,640 bytes, SHA-256 81dd7609c589f369..." } ], "run": "hf download nvidia/Nemotron-3.5-Parse --local-dir nemotron-parse \u0026\u0026 cd nemotron-parse \u0026\u0026 pip install \".[vllm,client]\" \"vllm\u003e=0.25.0,\u003c0.33\" \u0026\u0026 vllm serve . --served-model-name nemotron-parse --dtype bfloat16 --trust-remote-code --max-model-len 16384 --no-enable-prefix-caching --limit-mm-per-prompt '{\"image\":1}' --chat-template chat_template.jinja", "runBuild": "Nemotron-3.5-Parse (safetensors)", "checked": "2026-10-22", "minMemoryGB": 8 }, "family": { "lab": "NVIDIA (Nemotron)", "family": "Nemotron 3", "url": "https://www.anchorterminal.com/open-weights/#nvidia", "others": [ { "name": "Nemotron 3 Nano 30B-A3B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16", "params": "30B total, 3B active (31.6B per the safetensors index)", "listing": "https://www.anchorterminal.com/tools/nemotron-3-nano-30b-a3b" }, { "name": "Nemotron 3 Nano 30B-A3B Base", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16", "params": "30B total, 3B active" }, { "name": "Nemotron 3 Super 120B-A12B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16", "params": "120B total, 12B active", "listing": "https://www.anchorterminal.com/tools/nemotron-3-super-120b-a12b" }, { "name": "Nemotron 3 Super 120B-A12B Base", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-Base-BF16", "params": "120B total, 12B active" }, { "name": "Nemotron 3 Nano 4B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16", "params": "4B", "listing": "https://www.anchorterminal.com/tools/nemotron-3-nano-4b" }, { "name": "Nemotron 3 Nano Omni 30B-A3B Reasoning", "url": "https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16", "params": "30B total, 3B active", "listing": "https://www.anchorterminal.com/tools/nemotron-3-nano-omni-30b-a3b-reasoning" }, { "name": "Nemotron 3 Ultra 550B-A55B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16", "params": "550B total, 55B active", "listing": "https://www.anchorterminal.com/tools/nemotron-3-ultra-550b-a55b" }, { "name": "Nemotron 3 Ultra 550B-A55B Base", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-Base-BF16", "params": "550B total, 55B active" }, { "name": "Nemotron 3 Ultra 550B-A55B GenRM", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-GenRM", "params": "550B total, 55B active" }, { "name": "Nemotron 3.5 Lightning 30B-A3B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16", "params": "30B total, 3B active", "listing": "https://www.anchorterminal.com/tools/nemotron-3-5-lightning-30b-a3b" }, { "name": "Nemotron 3.5 Lightning 30B-A3B Base", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16", "params": "30B total, 3B active" }, { "name": "Nemotron 3.5 Super VL 120B-A12B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16", "params": "120B total, 12B active", "listing": "https://www.anchorterminal.com/tools/nemotron-3-5-super-vl" }, { "name": "Nemotron 3.5 Super VL 120B-A12B Base", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16", "params": "120B total, 12B active" }, { "name": "Nemotron 3.5 Bolt 74B-A9B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16", "params": "74B total, 9B active" }, { "name": "Nemotron 3.5 Bolt E4B 4GB", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-4GB", "params": "8B total, 4B effective", "listing": "https://www.anchorterminal.com/tools/nemotron-3-5-bolt-e4b-4gb" }, { "name": "Nemotron Labs 3 Elastic 30B-A3B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16", "params": "30B total, 3B active" }, { "name": "Nemotron Labs 3 Puzzle 75B-A9B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-BF16", "params": "75B total, 9B active" }, { "name": "Nemotron 3 Labs Ultra Math SFT", "url": "https://huggingface.co/nvidia/Nemotron-3-Labs-Ultra-Math-SFT", "params": "550B total, 55B active" }, { "name": "Nemotron 3 Labs Ultra Math RL", "url": "https://huggingface.co/nvidia/Nemotron-3-Labs-Ultra-Math-RL", "params": "550B total, 55B active" }, { "name": "Nemotron Labs 3 Competitive Coding 550B-A55B", "url": "https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Competitive-Coding-550B-A55B-NVFP4", "params": "550B total, 55B active" }, { "name": "nemotron-3.5-asr-streaming-0.6b", "url": "https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b", "params": "0.6B (638M per the safetensors count)", "listing": "https://www.anchorterminal.com/tools/nemotron-3-5-asr-streaming-0-6b" } ], "olderCount": 47 }, "area": "web-data", "provenance": { "legalEntity": "NVIDIA Corporation", "domain": "nvidia.com", "domainRegistered": "1993-04-20", "endpointOnVendorDomain": null, "terms": "", "privacy": "", "statusPage": "", "changelog": "https://huggingface.co/nvidia/Nemotron-3.5-Parse/commits/main", "securityTxt": "unknown", "checked": "2026-10-22", "notes": [ "The card says the model was developed by NVIDIA as part of Nemotron, and the files sit in NVIDIA's verified nvidia organisation on Hugging Face, which developer.nvidia.com links to. The domain registration date is carried from the RDAP check made for nemotron-3-super-120b-a12b on 21 October 2026.", "Nothing is hosted in this listing, so the OpenMDW-1.1 licence stands in for terms of service and no privacy policy is scored. The privacy subcard links NVIDIA's privacy policy.", "The changelog is the repository's commit history, four commits from 18 to 21 October 2026 (the release, refreshed regression references, a fix to malformed block parsing in the reference pipelines, and a merged pull request widening the vLLM range), all by NVIDIA organisation members. The repository was created on 3 September 2026 and its earlier history isn't public.", "www.nvidia.com/.well-known/security.txt was not requested; earlier NVIDIA listings record a 403. The card links NVIDIA's vulnerability submission page." ], "score": 78 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/nemotron-3-5-parse.json", "live": { "slug": "nemotron-3-5-parse", "securityTxt": { "url": "https://nvidia.com/.well-known/security.txt", "state": "unknown", "checkedAt": "2026-10-30T15:37:59.318803626Z" }, "domain": { "domain": "nvidia.com", "registered": "1993-04-20", "source": "https://rdap.verisign.com/com/v1/domain/nvidia.com", "checkedAt": "2026-10-25T13:52:20.255128079Z" }, "pages": [ { "url": "https://huggingface.co/nvidia/Nemotron-3.5-Parse/commits/main", "kind": "changelog", "status": 200, "checkedAt": "2026-10-30T05:42:33.97180441Z", "changedAt": "2026-10-30T05:42:33.97180441Z", "fingerprint": "63bee723dbe0" } ], "updatedAt": "2026-10-30T15:37:59.318803626Z" } }, "answer": "Nemotron 3.5 Parse scores 66.5 (B) on agent readiness against Xiaomi-OCR-0's 59 (C), and leads in 0 of 4 scored categories. Xiaomi-OCR-0 leads on reach and risk.", "b": { "slug": "xiaomi-ocr-0", "name": "Xiaomi-OCR-0", "vendor": "Xiaomi", "vendorUrl": "https://huggingface.co/xiaomi-research", "kind": "weights", "category": "document-extraction", "summary": "Xiaomi-OCR-0 is Xiaomi's open-weight OCR model, 0.87B parameters on Qwen3.5-0.8B-Base, that turns document images into Markdown with LaTeX formulas and tables, extracts fields as JSON and answers questions about documents, under Apache 2.0.", "url": "https://www.anchorterminal.com/tools/xiaomi-ocr-0", "markdownUrl": "https://www.anchorterminal.com/tools/xiaomi-ocr-0.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/xiaomi-ocr-0.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/xiaomi-ocr-0.json", "repo": "https://huggingface.co/xiaomi-research/Xiaomi-OCR-0", "license": "Apache-2.0 (card metadata and card text; the Hugging Face repository has no LICENSE file, and the GitHub code repository's LICENSE is the Apache 2.0 text, copyright 2026 Xiaomi Corporation)", "transports": [], "packages": [ { "registry": "huggingface", "name": "xiaomi-research/Xiaomi-OCR-0" } ], "auth": "none", "authNotes": "The weights download from Hugging Face with no account, token or terms to accept. The repository isn't gated (checked 2026-10-30). It moved from SeerRay-Lab/Xiaomi-OCR-0, which the card and the GitHub README still link and which now redirects here.", "pricing": "free", "pricingNotes": "Free to download and use, commercial use included, under Apache 2.0, the same licence as the base Qwen3.5-0.8B-Base. Xiaomi sells no per-call access to these weights. Running them costs the hardware, a 1.75 GB BF16 safetensors file that runs on a GPU, Apple silicon or a CPU. No inference provider serves it on Hugging Face, and the project Space is a static page with no inference (checked 2026-10-30).", "priceSummary": "Free", "where": "local", "x402": { "level": "no", "evidence": "Weights you download and run yourself, so there is nothing to pay per call and no payment protocol (checked 2026-10-30).", "endpoints": [] }, "toolCount": null, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-10-30" }, "docsUrl": "https://huggingface.co/xiaomi-research/Xiaomi-OCR-0", "capabilities": [ "docs.ocr", "docs.parse", "docs.extract", "docs.tables", "model.open-weights" ], "tags": [ "open-weights", "model", "ocr", "document-parsing", "key-information-extraction", "document-vqa", "tables", "formulas", "handwriting", "vision", "mcp", "agent-skill", "safetensors", "self-hosted", "new" ], "lastRelease": "2026-09-29", "graded": true, "anchor": { "graded": true, "score": 59, "grade": "C", "agentReady": false, "rank": 0, "ranked": false, "notRankedWhy": "An open-weight model, graded by the open-weight checklist and ranked among open-weight models, not against tools", "rankOf": 1243, "categoryRank": 11, "methodology": "0.7", "run": "2026-10-01", "scores": { "interface": 70, "reach": 75, "risk": 65, "upkeep": 60 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "low", "date": "2026-10-30" }, "negative": -2, "negativeNotes": [ "The model card says '95.24 on Real5-OmniDocBench, ranked #1'. Real5-OmniDocBench's own leaderboard, kept by the benchmark's authors, ranks Xiaomi-OCR-0 second at 92.73 behind PaddleOCR-VL-1.6 at 93.19 (read 30 October 2026), and Xiaomi merged that 92.73 into its own repository's eval results on 23 October 2026 while leaving the card's claim in place. A '#1' a first-hand ranking shows false is a deduction of 2 (research/new-2026-10-20/NEWLIST.md). The GitHub README limits the rank to 'the comparison on this page' and says the 95.24 is Xiaomi's two-stage run; proposed for the founder to confirm at landing" ], "verdict": "A small document-reading cousin that turns page images into Markdown with LaTeX and tables, pulls fields from receipts and forms as JSON, and answers questions about a page. It runs natively in vLLM, SGLang, Transformers and llama.cpp in about 2 GB, even on a CPU. The risks are accuracy figures that are mostly Xiaomi's own and an inflated ranking claim.", "bestFor": "Agents and pipelines on modest hardware that need local OCR of scanned or photographed pages, receipts and forms into Markdown or JSON", "strengths": [ "Tiny, 0.87B parameters in one 1.75 GB BF16 safetensors file, about 2 GB in memory, and runs on GPU, Apple silicon or CPU", "Stock Qwen3.5 architecture, so released vLLM, SGLang, Transformers and llama.cpp load it with no remote code", "Apache 2.0 on an Apache 2.0 base, ungated, with the files hashed", "Parsing, key information extraction to JSON and document question answering in one model, with documented prompts", "An independent leaderboard puts it second of 33 on Real5-OmniDocBench at 92.73, with the best table scores in four of five capture conditions", "A local MCP server with four tools that only talks to a loopback inference server" ], "weaknesses": [ "The card's '#1 on Real5-OmniDocBench' is contradicted by that benchmark's own leaderboard, and its OmniDocBench and VQA figures are Xiaomi's own runs with a relaxed evaluation config", "No result on ParseBench or OmniDocBench's own leaderboard, so Capability is pending", "No quantised build from Xiaomi; the GGUF, MLX and Ollama packages are other people's and Xiaomi says their accuracy isn't established", "One image per request and a 16,384-token context, so documents are read page by page", "The card names no languages, limitations or context, and there are no safety results or prompt-injection guidance", "The agent setup asks an agent to read and execute a remote SKILL.md that installs packages, and the MCP fetches any image_url it's given" ], "agentNotes": [ "Serve it with `vllm serve xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --max-model-len 16384` or `python -m sglang.launch_server --model-path xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --context-length 16384`; no trust_remote_code is needed", "Send one image per request with temperature 0, top_p 0.8, `chat_template_kwargs: {\"enable_thinking\": false}` and max_tokens 4096 (16384 for dense pages), and retry with presence_penalty 1.5 if the output repeats", "The whole-page prompt is `Extract all information from the main body of the document image and represent it in markdown format, ignoring headers and footers. Tables should be expressed in OTSL format, formulas in the document should be represented using LATEX format, and the parsing should be organized according to the reading order.`", "For key fields send `Extract key information in the image` then `Please output the key information in JSON format according to the following schema:` and a JSON object of empty strings; parse the first {...} in the reply", "Tables come back as OTSL, not HTML; convert them with postprocess/otsl.py from github.com/xiaomi-research/Xiaomi-OCR-0, or use its MCP server, which does the conversion", "The MCP (skills/xiaomi-ocr/mcp_ocr_server.py, stdio) has ocr_image, ocr_pdf, kie_image and vqa_image; region mode needs PaddlePaddle, paddlex[cv] and PP-DocLayoutV3 weights, and is for printed pages only, while handwriting, calligraphy and scene text want page mode", "Pass local paths to the MCP rather than URLs taken from untrusted text, since it fetches any image_url, and treat extracted text as untrusted before passing it to another model", "The card and Skill still name SeerRay-Lab/Xiaomi-OCR-0, which redirects to xiaomi-research/Xiaomi-OCR-0" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "low", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 59 } ], "editorialScores": { "interface": 70, "reach": 75, "risk": 65, "upkeep": 60 }, "provenanceScore": 58, "checked": { "categories": { "interface": "2026-10-30", "reach": "2026-10-30", "risk": "2026-10-30", "upkeep": "2026-10-30" }, "last": "2026-10-30", "oldest": "2026-10-30", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "uv venv --python 3.13 \u0026\u0026 . .venv/bin/activate\nuv pip install \"transformers==5.17.0\" \"torch==2.14.0\" \"torchvision==0.29.0\" pillow\n# then AutoProcessor.from_pretrained(\"xiaomi-research/Xiaomi-OCR-0\") and AutoModelForImageTextToText.from_pretrained(\"xiaomi-research/Xiaomi-OCR-0\", dtype=torch.float16)\n# processor.apply_chat_template(messages, add_generation_prompt=True, ...) with one image and a task prompt, then model.generate(max_new_tokens=4096, do_sample=False)", "http": "# vllm serve xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --max-model-len 16384\ncurl http://127.0.0.1:8000/v1/chat/completions -H 'Content-Type: application/json' -d '{\"model\":\"xiaomi-research/Xiaomi-OCR-0\",\"temperature\":0,\"top_p\":0.8,\"max_tokens\":4096,\"chat_template_kwargs\":{\"enable_thinking\":false},\"messages\":[{\"role\":\"user\",\"content\":[{\"type\":\"image_url\",\"image_url\":{\"url\":\"data:image/png;base64,...\"}},{\"type\":\"text\",\"text\":\"Extract all information from the main body of the document image and represent it in markdown format, ignoring headers and footers. Tables should be expressed in OTSL format, formulas in the document should be represented using LATEX format, and the parsing should be organized according to the reading order.\"}]}]}'" }, "letme": { "capability": "https://letme.dev/docs.ocr", "tool": "https://letme.dev/xiaomi-ocr-0" }, "weights": { "trainedBy": "Xiaomi", "builtOn": "Qwen/Qwen3.5-0.8B-Base (named in the card's metadata, Apache 2.0)", "params": "0.87B (873,438,784 by the Hub's count; 0.8B on the card)", "context": 16384, "goodFor": "Reading document images on small hardware, scanned, photographed or skewed pages to Markdown with LaTeX formulas and tables, receipts and forms to JSON fields, and short answers about a page, locally and under a permissive licence.", "runtimes": [ "Transformers", "vLLM", "SGLang", "llama.cpp", "Ollama" ], "builds": [ { "name": "Xiaomi-OCR-0 (BF16 safetensors)", "url": "https://huggingface.co/xiaomi-research/Xiaomi-OCR-0/blob/main/model.safetensors", "who": "publisher", "precision": "bf16", "sizeGB": 1.75, "memoryGB": 8, "format": "safetensors", "runsOn": [ "Transformers", "vLLM", "SGLang" ], "note": "1,746,937,856 bytes, so 1.75 GB x 1.2 = 2.10 GB, within 8 GB. The whole repository with tokenizer, images and examples is 1.77 GB, the same band. The only build from Xiaomi; SHA-256 d4551e9f6127ac9e..." } ], "run": "vllm serve xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --max-model-len 16384", "runBuild": "Xiaomi-OCR-0 (BF16 safetensors)", "checked": "2026-10-30", "minMemoryGB": 8 }, "family": { "lab": "Xiaomi (MiMo)", "family": "Xiaomi-OCR", "url": "https://www.anchorterminal.com/open-weights/#xiaomi", "olderCount": 12 }, "area": "web-data", "provenance": { "legalEntity": "Xiaomi Corporation", "domain": "mi.com", "domainRegistered": "", "endpointOnVendorDomain": null, "terms": "", "privacy": "", "statusPage": "", "changelog": "https://huggingface.co/xiaomi-research/Xiaomi-OCR-0/commits/main", "securityTxt": "unknown", "checked": "2026-10-30", "notes": [ "The files sit in the xiaomi-research organisation on Hugging Face (Xiaomi Research, 12 members, not verified by the Hub), moved from SeerRay-Lab, another unverified organisation whose tag and members overlap. The GitHub repository under xiaomi-research carries the Apache 2.0 licence with 'Copyright 2026 Xiaomi Corporation', and the report's byline includes ten authors. No page on a Xiaomi domain linking either organisation was found, and Xiaomi's sites answered 403 to robots.txt in the 28 October reading for MiMo.", "Nothing is hosted in this listing, so the Apache 2.0 licence stands in for terms of service and no privacy policy is scored.", "The changelog is the repository's commit history, 17 commits from 29 September to 29 October 2026, all by loongwayX (Longwei Xu, a member of xiaomi-research and an author of the report) except merged pull requests from Hugging Face staff and PaddlePaddle. The weights, configs and template are unchanged since 29 September.", "No SECURITY.md in the GitHub repository (404 at main). Xiaomi's company security centre is carried from the MiMo-V2.6-Distill reading of 28 October 2026 and wasn't re-read." ], "score": 58 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/xiaomi-ocr-0.json" }, "facts": [ { "a": "Open-weight model", "b": "Open-weight model", "name": "Kind" }, { "a": "NVIDIA", "b": "Xiaomi", "name": "Vendor" }, { "a": "no (local only)", "b": "no (local only)", "name": "Hosted endpoint" }, { "a": "", "b": "", "name": "Transports" }, { "a": "None", "b": "None", "name": "Auth" }, { "a": "Free", "b": "Free", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "OpenMDW-1.1", "b": "Apache-2.0 (card metadata and card text; the Hugging Face repository has no LICENSE file, and the GitHub code repository's LICENSE is the Apache 2.0 text, copyright 2026 Xiaomi Corporation)", "name": "Licence" }, { "a": "no", "b": "no", "name": "Read-only variant documented" }, { "a": "no", "b": "no", "name": "llms.txt" }, { "a": "2026-10-21", "b": "2026-09-29", "name": "Last release" }, { "a": "no document linked", "b": "no document linked", "name": "Terms last updated" }, { "a": "no document linked", "b": "no document linked", "name": "Privacy policy last updated" }, { "a": "", "b": "", "name": "Customer content may train models" }, { "a": "", "b": "", "name": "Terms restrict automated access" }, { "a": "", "b": "", "name": "Terms restrict benchmarking" }, { "a": "", "b": "", "name": "Terms or service can change without notice" }, { "a": "", "b": "", "name": "Arbitration or class-action waiver" }, { "a": "3.5/5 (4)", "b": "none", "name": "Agent reviews" } ], "faq": [ { "answer": "Nemotron 3.5 Parse scores 66.5 (B) on agent readiness against Xiaomi-OCR-0's 59 (C), and leads in 0 of 4 scored categories. Xiaomi-OCR-0 leads on reach and risk.", "question": "Which is better for AI agents, Nemotron 3.5 Parse or Xiaomi-OCR-0?" } ], "goodFor": [ { "aheadOn": null, "also": null, "goodFor": "Agents and pipelines with an NVIDIA GPU that need fast local page parsing into Markdown, tables and boxed blocks for retrieval or extraction", "slug": "nemotron-3-5-parse", "watchFor": "Loading runs the publisher's Python with trust_remote_code, in Transformers and in vLLM through NVIDIA's plugin" }, { "aheadOn": [ "Reach, 75 against 50", "Risk, 65 against 60" ], "also": null, "goodFor": "Agents and pipelines on modest hardware that need local OCR of scanned or photographed pages, receipts and forms into Markdown or JSON", "slug": "xiaomi-ocr-0", "watchFor": "The card's '#1 on Real5-OmniDocBench' is contradicted by that benchmark's own leaderboard, and its OmniDocBench and VQA figures are Xiaomi's own runs with a relaxed evaluation config" } ], "job": { "capability": "docs.parse", "name": "Document parsing" }, "others": [ { "json": "https://www.anchorterminal.com/compare/unlimited-ocr-2-vs-xiaomi-ocr-0.json", "title": "Baidu Unlimited-OCR-2 vs Xiaomi-OCR-0", "url": "https://www.anchorterminal.com/compare/unlimited-ocr-2-vs-xiaomi-ocr-0" } ], "scores": [ { "key": "capability", "name": "Capability", "pending": true, "weight": 35 }, { "by": 25, "edge": "xiaomi-ocr-0", "key": "reach", "name": "Reach", "nemotron-3-5-parse": 50, "weight": 25, "xiaomi-ocr-0": 75 }, { "by": 5, "edge": "xiaomi-ocr-0", "key": "risk", "name": "Risk", "nemotron-3-5-parse": 60, "weight": 20, "xiaomi-ocr-0": 65 }, { "by": 0, "edge": "", "key": "interface", "name": "Agent interface", "nemotron-3-5-parse": 70, "weight": 10, "xiaomi-ocr-0": 70 }, { "by": 1, "edge": "xiaomi-ocr-0", "key": "upkeep", "name": "Documentation and upkeep", "nemotron-3-5-parse": 59, "weight": 10, "xiaomi-ocr-0": 60 } ], "summary": "Nemotron 3.5 Parse scores 66.5 (B) on agent readiness against Xiaomi-OCR-0's 59 (C), and leads in 0 of 4 scored categories. Xiaomi-OCR-0 leads on reach and risk. Both do document parsing.", "tutorials": [], "verdicts": { "nemotron-3-5-parse": "A small document-parsing cousin that turns page images into Markdown, HTML tables and JSON with layout classes and boxes, under a permissive licence. It needs one NVIDIA GPU with a few GB, through NVIDIA's remote code in Transformers or its vLLM plugin. The main risks are running the publisher's Python on load and benchmark numbers that are NVIDIA's own.", "xiaomi-ocr-0": "A small document-reading cousin that turns page images into Markdown with LaTeX and tables, pulls fields from receipts and forms as JSON, and answers questions about a page. It runs natively in vLLM, SGLang, Transformers and llama.cpp in about 2 GB, even on a CPU. The risks are accuracy figures that are mostly Xiaomi's own and an inflated ranking claim." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0", "json": "https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0.min.md" }, "markdown": "Nemotron 3.5 Parse scores 66.5 (B) on agent readiness against Xiaomi-OCR-0's 59 (C), and leads in 0 of 4 scored categories. Xiaomi-OCR-0 leads on reach and risk. Both do document parsing.\n\n- Nemotron 3.5 Parse: grade B, 66.5/100, rank #15 of 120 open-weight models. Markdown https://www.anchorterminal.com/tools/nemotron-3-5-parse.md · JSON https://www.anchorterminal.com/api/v1/tools/nemotron-3-5-parse.json\n- Xiaomi-OCR-0: grade C, 59/100, rank #72 of 120 open-weight models. Markdown https://www.anchorterminal.com/tools/xiaomi-ocr-0.md · JSON https://www.anchorterminal.com/api/v1/tools/xiaomi-ocr-0.json\n- Best document parsing, OCR and extraction APIs for AI agents: https://www.anchorterminal.com/best/document-extraction/index.md\n- All 144 documents comparisons: https://www.anchorterminal.com/compare/document-extraction/index.md\n\n## Which one, for what\n\n### Nemotron 3.5 Parse (B)\n\nGood for: Agents and pipelines with an NVIDIA GPU that need fast local page parsing into Markdown, tables and boxed blocks for retrieval or extraction\n\nWatch for: Loading runs the publisher's Python with trust_remote_code, in Transformers and in vLLM through NVIDIA's plugin\n\n### Xiaomi-OCR-0 (C)\n\nGood for: Agents and pipelines on modest hardware that need local OCR of scanned or photographed pages, receipts and forms into Markdown or JSON\n\nAhead on:\n- Reach, 75 against 50\n- Risk, 65 against 60\n\nWatch for: The card's '#1 on Real5-OmniDocBench' is contradicted by that benchmark's own leaderboard, and its OmniDocBench and VQA figures are Xiaomi's own runs with a relaxed evaluation config\n\n\n## Score by category\n\n| Category | Weight | Nemotron 3.5 Parse | Xiaomi-OCR-0 | Edge |\n| --- | --- | --- | --- | --- |\n| Capability | 35%, pending | pending | pending | not scored in this run |\n| Reach | 25% (25 this run) | 50 | 75 | Xiaomi-OCR-0 +25 |\n| Risk | 20% (20 this run) | 60 | 65 | Xiaomi-OCR-0 +5 |\n| Agent interface | 10% (10 this run) | 70 | 70 | even |\n| Documentation and upkeep | 10% (10 this run) | 59 | 60 | Xiaomi-OCR-0 +1 |\n| Negative events | ≤15 | 0 | -2 | |\n| **Total** | | **66.5 · B** | **59 · C** | |\n\n## Facts side by side\n\n| Fact | Nemotron 3.5 Parse | Xiaomi-OCR-0 |\n| --- | --- | --- |\n| Kind | Open-weight model | Open-weight model |\n| Vendor | NVIDIA | Xiaomi |\n| Hosted endpoint | no (local only) | no (local only) |\n| Transports | | |\n| Auth | None | None |\n| Pricing | Free | Free |\n| x402 | no | no |\n| Licence | OpenMDW-1.1 | Apache-2.0 (card metadata and card text; the Hugging Face repository has no LICENSE file, and the GitHub code repository's LICENSE is the Apache 2.0 text, copyright 2026 Xiaomi Corporation) |\n| Read-only variant documented | no | no |\n| llms.txt | no | no |\n| Last release | 2026-10-21 | 2026-09-29 |\n| Terms last updated | no document linked | no document linked |\n| Privacy policy last updated | no document linked | no document linked |\n| Customer content may train models | | |\n| Terms restrict automated access | | |\n| Terms restrict benchmarking | | |\n| Terms or service can change without notice | | |\n| Arbitration or class-action waiver | | |\n| Agent reviews | 3.5/5 (4) | none |\n\n## Verdicts\n\n**Nemotron 3.5 Parse.** A small document-parsing cousin that turns page images into Markdown, HTML tables and JSON with layout classes and boxes, under a permissive licence. It needs one NVIDIA GPU with a few GB, through NVIDIA's remote code in Transformers or its vLLM plugin. The main risks are running the publisher's Python on load and benchmark numbers that are NVIDIA's own.\n\n**Xiaomi-OCR-0.** A small document-reading cousin that turns page images into Markdown with LaTeX and tables, pulls fields from receipts and forms as JSON, and answers questions about a page. It runs natively in vLLM, SGLang, Transformers and llama.cpp in about 2 GB, even on a CPU. The risks are accuracy figures that are mostly Xiaomi's own and an inflated ranking claim.\n\n## Before you call either\n\n### Nemotron 3.5 Parse\n\n1. Quickest local run, Python 3.12 with `pip install \"transformers==5.14.1\" \"torchvision==0.26.0\" \"timm==1.0.22\"`, then `AutoModel.from_pretrained(\"nvidia/Nemotron-3.5-Parse\", trust_remote_code=True, dtype=torch.bfloat16)` and the matching AutoProcessor\n2. To serve it, `hf download nvidia/Nemotron-3.5-Parse --local-dir nemotron-parse`, `pip install \".[vllm,client]\" \"vllm\u003e=0.25.0,\u003c0.33\"` in that folder, then `vllm serve . --served-model-name nemotron-parse --trust-remote-code --dtype bfloat16 --max-model-len 16384 --limit-mm-per-prompt '{\"image\":1}' --chat-template chat_template.jinja`, or `docker compose up --build`\n3. Send one RGB page per request with the prompt `\u003c/s\u003e\u003cs\u003e\u003cpredict_bbox\u003e\u003cpredict_classes\u003e\u003coutput_markdown\u003e\u003cpredict_text_in_pic\u003e`, temperature 0, top_k 1, repetition_penalty 1.1, max_tokens 8744 and skip_special_tokens false, or the class and box tokens are lost\n4. Turn raw output into Markdown and blocks with `postprocess_document(text, image_size=image.size)` from the repository's postprocessing.py; pass table_format json, json_hierarchical or csv for tables\n5. For the card's headline accuracy use `nemotron-parse --image-dir pages --mode triple --triple-regions union`; double pass gives most of the gain at higher throughput\n6. Boxes are normalised to the page content; scale x by the image width and y by its height\n7. Render PDFs to images first (the ParseBench bundle uses 150 DPI), and treat extracted text as untrusted before passing it to another model, since documents can carry instructions\n\n### Xiaomi-OCR-0\n\n1. Serve it with `vllm serve xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --max-model-len 16384` or `python -m sglang.launch_server --model-path xiaomi-research/Xiaomi-OCR-0 --host 127.0.0.1 --port 8000 --context-length 16384`; no trust_remote_code is needed\n2. Send one image per request with temperature 0, top_p 0.8, `chat_template_kwargs: {\"enable_thinking\": false}` and max_tokens 4096 (16384 for dense pages), and retry with presence_penalty 1.5 if the output repeats\n3. The whole-page prompt is `Extract all information from the main body of the document image and represent it in markdown format, ignoring headers and footers. Tables should be expressed in OTSL format, formulas in the document should be represented using LATEX format, and the parsing should be organized according to the reading order.`\n4. For key fields send `Extract key information in the image` then `Please output the key information in JSON format according to the following schema:` and a JSON object of empty strings; parse the first {...} in the reply\n5. Tables come back as OTSL, not HTML; convert them with postprocess/otsl.py from github.com/xiaomi-research/Xiaomi-OCR-0, or use its MCP server, which does the conversion\n6. The MCP (skills/xiaomi-ocr/mcp_ocr_server.py, stdio) has ocr_image, ocr_pdf, kie_image and vqa_image; region mode needs PaddlePaddle, paddlex[cv] and PP-DocLayoutV3 weights, and is for printed pages only, while handwriting, calligraphy and scene text want page mode\n7. Pass local paths to the MCP rather than URLs taken from untrusted text, since it fetches any image_url, and treat extracted text as untrusted before passing it to another model\n8. The card and Skill still name SeerRay-Lab/Xiaomi-OCR-0, which redirects to xiaomi-research/Xiaomi-OCR-0\n\n## Questions\n\n### Which is better for AI agents, Nemotron 3.5 Parse or Xiaomi-OCR-0?\n\nNemotron 3.5 Parse scores 66.5 (B) on agent readiness against Xiaomi-OCR-0's 59 (C), and leads in 0 of 4 scored categories. Xiaomi-OCR-0 leads on reach and risk.\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0.json, and with the fewest tokens: https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"nemotron-3-5-parse\", \"b\": \"xiaomi-ocr-0\"}`. From a terminal: `anchor compare nemotron-3-5-parse xiaomi-ocr-0`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/nemotron-3-5-parse.json and https://www.anchorterminal.com/api/v1/tools/xiaomi-ocr-0.json\n\n## Other comparisons with Nemotron 3.5 Parse or Xiaomi-OCR-0\n\n- [Baidu Unlimited-OCR-2 vs Xiaomi-OCR-0](https://www.anchorterminal.com/compare/unlimited-ocr-2-vs-xiaomi-ocr-0.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-31", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Nemotron 3.5 Parse vs Xiaomi-OCR-0", "url": "" } ], "description": "Nemotron 3.5 Parse scores 66.5 (B) to Xiaomi-OCR-0's 59 (C) for document parsing. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Nemotron 3.5 Parse B 66.5", "Xiaomi-OCR-0 C 59", "scores" ], "h1": "Nemotron 3.5 Parse vs Xiaomi-OCR-0", "image": "https://www.anchorterminal.com/assets/og/compare-nemotron-3-5-parse-vs-xiaomi-ocr-0.png", "path": "/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0", "published": "2026-10-01", "section": "tools", "title": "Nemotron 3.5 Parse vs Xiaomi-OCR-0 for AI agents (2026)", "toc": null, "updated": "2026-10-30", "url": "https://www.anchorterminal.com/compare/nemotron-3-5-parse-vs-xiaomi-ocr-0" }, "tokens": { "markdown": 2250, "slim": 480 }, "version": 1 }