{"data":{"slug":"hkuds-lightrag","name":"LightRAG","tagline":"Simple and Fast Retrieval-Augmented Generation","github_url":"https://github.com/HKUDS/LightRAG","owner":"HKUDS","repo":"LightRAG","owner_avatar_url":"https://avatars.githubusercontent.com/u/118165258?v=4","primary_language":"Python","stars":39743,"forks":5589,"topics":["docling","genai","gpt","graphrag","knowledge-graph","large-language-models","llm","mineru","rag","ragas","retrieval-augmented-generation"],"archived":false,"github_pushed_at":"2026-09-18T08:26:33+00:00","maintenance_label":"Very active","stars_delta_30d":848,"url":"https://www.graphcanon.com/tools/hkuds-lightrag","markdown_url":"https://www.graphcanon.com/tools/hkuds-lightrag.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/hkuds-lightrag","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=hkuds-lightrag","description":"[EMNLP2025] LightRAG: Simple and Fast Retrieval-Augmented Generation","homepage_url":"https://arxiv.org/abs/2410.05779","license":"MIT","open_issues":249,"watchers":301,"ai_summary":"LightRAG is a retrieval-augmented generation tool that focuses on simplicity and speed, designed for efficient knowledge graph augmentation in large language models.","readme_excerpt":"## Installation\n\n**💡 Using uv for Package Management**: This project uses [uv](https://docs.astral.sh/uv/) for fast and reliable Python package management. Install uv first: `curl -LsSf https://astral.sh/uv/install.sh | sh` (Unix/macOS) or `powershell -c \"irm https://astral.sh/uv/install.ps1 | iex\"` (Windows)\n\n> **Note**: You can also use pip if you prefer, but uv is recommended for better performance and more reliable dependency management.\n>\n> **📦 Offline Deployment**: For offline or air-gapped environments, see the [Offline Deployment Guide](./docs/OfflineDeployment.md) for instructions on pre-installing all dependencies and cache files.\n\n---\n\n### Install LightRAG Server\n\n* Install from PyPI\n\n```bash\n\n---\n\n### Install LightRAG Server as tool using uv (recommended)\nuv tool install \"lightrag-hku[api]\"\n\n---\n\n# Debian / Ubuntu (the official Docker image already includes this)\nsudo apt-get install -y libcairo2\n\n---\n\n# Windows: install the GTK3 runtime, which bundles libcairo-2.dll\n```\n\nDeployments that never process markdown/textpack documents with embedded SVGs can ignore the startup warning.\n\n---\n\n### Install LightRAG SDK\n\n* Install from source code\n\n```bash\ncd LightRAG\n\n---\n\n# 或: pip install -e .\n```\n\n* Install from PyPI\n\n```bash\nuv pip install lightrag-hku","github_created_at":"2024-10-02T11:57:54+00:00","created_at":"2026-07-07T17:31:41.413652+00:00","updated_at":"2026-09-20T04:32:29.872873+00:00","categories":[{"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"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"docling","name":"docling"},{"slug":"genai","name":"genai"},{"slug":"gpt","name":"gpt"},{"slug":"graphrag","name":"graphrag"},{"slug":"knowledge-graph","name":"knowledge-graph"},{"slug":"large-language-models","name":"large-language-models"},{"slug":"llm","name":"llm"},{"slug":"mineru","name":"mineru"}],"trust":{"provenance":{"is_fork":false,"github_id":866513204,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:40:05.845Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":7,"days_since_push":0,"last_release_at":"2026-09-02T07:23:55Z","stars_delta_30d":848,"open_issues_delta_30d":22},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-09-18T08:40:08.317Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:40:04.705Z"},"deploy":{"source":"dockerfile:Dockerfile","self_host":true,"observed_at":"2026-09-18T08:40:04.705Z","managed_saas":false},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-09-18T08:40:04.705Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-18T08:40:04.705Z"},"has_docker":{"value":true,"source":"dockerfile:Dockerfile","observed_at":"2026-09-18T08:40:04.705Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-09-18T08:40:04.705Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need a tool that prioritizes simplicity and speed in retrieval-augmented generation tasks.","For projects that require efficient knowledge graph augmentation in large language models.","If you prefer using uv for package management to ensure better performance and reliable dependency management.","In environments where offline deployment is necessary, as LightRAG provides a guide for pre-installing all dependencies and cache files."],"when_not_to_use":["If your project does not require efficient knowledge graph augmentation or if speed and simplicity are not primary concerns.","When you prefer not to use uv for package management and are satisfied with traditional methods like pip.","In scenarios where the installation of additional libraries such as libcairo2 for handling SVGs in markdown/textpack documents is not feasible or desired.","If your deployment environment strictly prohibits the use of external package managers like uv."],"source":"enrich:decision_facts","observed_at":"2026-09-18T11:41:08.374Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"LightRAG is a retrieval-augmented generation tool that emphasizes simplicity and speed, ideal for efficient knowledge graph augmentation in large language models."}]}}