{"data":{"slug":"aiptimizer-turboocr","name":"TurboOCR","tagline":"Fast GPU-based OCR server for document parsing","github_url":"https://github.com/aiptimizer/TurboOCR","owner":"aiptimizer","repo":"TurboOCR","owner_avatar_url":"https://avatars.githubusercontent.com/u/274521723?v=4","primary_language":"C++","stars":1078,"forks":111,"topics":["document-ai","document-parsing","easyocr","fastapi","fp16","gpu-ocr","grpc","inference-server","nvidia","ocr","paddleocr","pdf-extraction","qwen-vl","rag","tensorrt","text-detection","text-recognition"],"archived":false,"github_pushed_at":"2026-09-08T14:32:18+00:00","maintenance_label":"Active","stars_delta_30d":92,"url":"https://www.graphcanon.com/tools/aiptimizer-turboocr","markdown_url":"https://www.graphcanon.com/tools/aiptimizer-turboocr.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/aiptimizer-turboocr","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=aiptimizer-turboocr","description":"TurboOCR, >200 img/s OmnidocBench. TensorRT FP16, PP-OCRv6, HTTP + gRPC","homepage_url":"https://turboocr.com","license":"MIT","open_issues":1,"watchers":10,"ai_summary":"TurboOCR is an OCR server optimized for fast text detection and recognition on GPU using PP-OCRv5 with TensorRT FP16 acceleration.","readme_excerpt":"## Quick Start\n\n**Requirements:** Linux, NVIDIA driver 595+, Turing or newer GPU (RTX 20-series / GTX 16-series+). Plan for ~4 GB VRAM text-only and ~8 GB for the full pipeline (layout + tables + formulas); each extra `PIPELINE_POOL_SIZE` replica adds roughly another full set, so lower it on smaller cards.\n\n```bash\ndocker run --gpus all -p 8000:8000 -p 50051:50051 \\\n  -v trt-cache:/home/ocr/.cache/turbo-ocr \\\n  ghcr.io/aiptimizer/turboocr:latest\n```\n\nFirst startup builds TensorRT engines from ONNX. This takes about 90 seconds on a\n5090 GPU and up to an hour on older ones. Set `TRT_OPT_LEVEL=3` to cut build time\n3 to 5x with a small speed regression. The named volume caches the engines, so\nsubsequent starts are instant. During the build, requests return a connection\nrefused error from nginx until the backend is ready. nginx (port 8000)\nreverse-proxies to Drogon (port 8080), and both start automatically.\n\n```bash\ncurl -X POST http://localhost:8000/ocr/raw \\\n  --data-binary @document.png -H \"Content-Type: image/png\"\n```\n\n```json\n{\"results\": [{\"text\": \"Invoice Total\", \"confidence\": 0.97, \"bounding_box\": [[42,10],[210,10],[210,38],[42,38]]}]}\n```\n\n---\n\n# Docker (recommended)\ndocker build -f docker/Dockerfile.gpu -t turboocr .\ndocker run --gpus all -p 8000:8000 -p 50051:50051 \\\n  -v trt-cache:/home/ocr/.cache/turbo-ocr turboocr\n\n---\n\n## License\n\nMIT. See [LICENSE](LICENSE).\n\n<p align=\"center\">\n  <a href=\"https://github.com/aiptimizer/TurboOCR\"><strong>⭐ Star TurboOCR on GitHub</strong></a><br>\n  <sub>Sponsored by <a href=\"https://miruiq.com\"><strong>Miruiq</strong></a> — AI-powered data extraction from PDFs and documents — and <a href=\"https://diaiq.com\"><strong>DiaIQ</strong></a>.</sub>\n</p>","github_created_at":"2026-03-20T10:42:34+00:00","created_at":"2026-07-15T11:19:31.508532+00:00","updated_at":"2026-09-20T05:19:52.353101+00:00","categories":[{"slug":"computer-vision","name":"Computer Vision","url":"https://www.graphcanon.com/categories/computer-vision","markdown_url":"https://www.graphcanon.com/categories/computer-vision.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/computer-vision"},{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"}],"tags":[{"slug":"document-ai","name":"document-ai"},{"slug":"document-parsing","name":"document-parsing"},{"slug":"gpu-ocr","name":"gpu-ocr"},{"slug":"grpc","name":"grpc"},{"slug":"text-detection","name":"text-detection"},{"slug":"text-recognition","name":"text-recognition"}],"trust":{"provenance":{"is_fork":false,"github_id":1187099783,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T05:19:50.581Z","maintenance":{"label":"Active","score":82,"methodology":"github_public_v1","releases_90d":15,"days_since_push":11,"last_release_at":"2026-09-08T15:23:09Z","stars_delta_30d":92,"open_issues_delta_30d":0},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-15T11:19:34.062Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-20T05:19:51.595Z"},"languages":{"value":["c++"],"source":"github.language","observed_at":"2026-09-20T05:19:51.595Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-20T05:19:51.595Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need fast text detection and recognition on GPU-enabled systems","For document parsing tasks where speed is critical and NVIDIA GPUs are available"],"when_not_to_use":["If your infrastructure lacks a supported NVIDIA GPU or has less than required VRAM","In scenarios where the initial TensorRT engine build time cannot be accommodated"],"source":"enrich:decision_facts","observed_at":"2026-07-17T11:37:00.600Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"TurboOCR, optimized for GPUs, delivers high-speed OCR with support for TensorRT FP16 acceleration and PP-OCRv5 model."}]}}