{"data":{"slug":"startrail-org-pixelrag","name":"PixelRAG","tagline":"Scalable pixel-native search for multimodal data","github_url":"https://github.com/StarTrail-org/PixelRAG","owner":"StarTrail-org","repo":"PixelRAG","owner_avatar_url":"https://avatars.githubusercontent.com/u/288858980?v=4","primary_language":"Python","stars":9586,"forks":817,"topics":["agent","ai","memory","multimodal","rag","search","searchengine","vision","vlm"],"archived":false,"github_pushed_at":"2026-07-31T08:52:00+00:00","maintenance_label":"Active","stars_delta_30d":2777,"url":"https://www.graphcanon.com/tools/startrail-org-pixelrag","markdown_url":"https://www.graphcanon.com/tools/startrail-org-pixelrag.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/startrail-org-pixelrag","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=startrail-org-pixelrag","description":"The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/","homepage_url":"https://arxiv.org/pdf/2606.28344","license":"Apache-2.0","open_issues":24,"watchers":36,"ai_summary":"PixelRAG transforms PDFs into searchable tiles, enabling scalable and efficient multimodal data retrieval.","readme_excerpt":"# PDF → tiles (requires poppler; install the pdf extra: pip install 'pixelrag[pdf]')\ncurl -sL -o paper.pdf https://arxiv.org/pdf/2503.09516\npixelshot paper.pdf -o ./tiles --dpi 200\n\n---\n\n# Start one locally with: docker run -p 6333:6333 qdrant/qdrant\npixelrag build-index --embeddings-dir ./embeddings --output-dir ./index \\\n    --backend qdrant --qdrant-url http://localhost:6333 --collection pixelrag \\\n    --qdrant-quantization-config ./quantization.json","github_created_at":"2026-05-29T07:25:40+00:00","created_at":"2026-07-07T17:37:46.835156+00:00","updated_at":"2026-08-18T18:00:50.226085+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":"data-retrieval","name":"Data & Retrieval","url":"https://www.graphcanon.com/categories/data-retrieval","markdown_url":"https://www.graphcanon.com/categories/data-retrieval.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/data-retrieval"}],"tags":[{"slug":"multimodal","name":"multimodal"},{"slug":"searchengine","name":"searchengine"},{"slug":"vision","name":"vision"},{"slug":"vlm","name":"vlm"}],"trust":{"provenance":{"is_fork":false,"github_id":1253138201,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-18T18:00:47.868Z","maintenance":{"label":"Active","score":82,"methodology":"github_public_v1","releases_90d":4,"days_since_push":18,"last_release_at":"2026-07-16T10:44:48Z","stars_delta_30d":2777,"open_issues_delta_30d":13},"security_summary":{"status":"ok","scanner":"osv@v1","low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:10:01.918Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"mcp":{"source":"repo_scan","observed_at":"2026-08-18T18:00:49.489Z","server_manifest":false},"scan":{"source":"repo_scan","observed_at":"2026-08-18T18:00:49.489Z"},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-08-18T18:00:49.489Z"},"languages":{"value":["python","javascript"],"source":"github.language+package.json+pyproject.toml","observed_at":"2026-08-18T18:00:49.489Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-08-18T18:00:49.489Z"}},"decision_facts":{"hosting":null,"pricing":{"model":"unknown","summary":"The pricing information is not available from the current repository data."},"requirements":{"notes":["Requires installation of 'poppler' to handle PDF files effectively. Use `pip install 'pixelrag[pdf]'` for complete setup."]},"constraints":{"pricing_model":"unknown"},"when_to_use":["When your application requires scalable pixel-native search capabilities for multimodal data, particularly from PDF documents","If you are working with dense text and graphical content within PDF files where traditional web parsing is insufficient or inefficient"],"when_not_to_use":["For tasks that do not require the conversion of textual or graphically rich content into searchable formats, as PixelRAG is PDF-centric and might not offer value in other data contexts","If you are dealing exclusively with text-based search and your data format doesn't include substantial graphical elements; another tool might be more efficient"],"source":"enrich:decision_facts","observed_at":"2026-07-12T11:36:43.610Z"},"constraint_facets":{"pricing_model":"unknown"},"decision_summary":[{"label":"Pricing","value":"unknown - The pricing information is not available from the current repository data."},{"label":"Requirements","value":"Requires installation of 'poppler' to handle PDF files effectively. Use `pip install 'pixelrag[pdf]'` for complete setup."},{"label":"Adopt for","value":"PixelRAG is a Python-based tool that specializes in transforming PDFs into searchable image tiles, enabling efficient and scalable multimodal data retrieval."},{"label":"License detail","value":"PixelRAG operates under an Apache 2.0 license, which allows for both commercial use and modification of the code."}]}}