{"id":13489365,"url":"https://github.com/infiniflow/infinity","last_synced_at":"2026-04-02T12:01:53.763Z","repository":{"id":213494002,"uuid":"515204860","full_name":"infiniflow/infinity","owner":"infiniflow","description":"The AI-native database built for LLM applications, providing incredibly fast hybrid search of dense vector, sparse vector, tensor (multi-vector), and full-text","archived":false,"fork":false,"pushed_at":"2025-05-12T07:46:36.000Z","size":65170,"stargazers_count":3590,"open_issues_count":108,"forks_count":341,"subscribers_count":39,"default_branch":"main","last_synced_at":"2025-05-12T13:17:59.886Z","etag":null,"topics":["ai-native","approximate-nearest-neighbor-search","bm25","cpp20","cpp20-modules","embedding","full-text-search","hnsw","hybrid-search","information-retrival","nearest-neighbor-search","rag","search-engine","tensor-database","vector","vector-database","vector-search","vectordatabase"],"latest_commit_sha":null,"homepage":"https://infiniflow.org","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/infiniflow.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2022-07-18T13:52:38.000Z","updated_at":"2025-05-12T11:16:09.000Z","dependencies_parsed_at":"2025-04-23T18:48:51.230Z","dependency_job_id":"901f443d-17c3-456d-afb5-c3704bf494ad","html_url":"https://github.com/infiniflow/infinity","commit_stats":null,"previous_names":["infiniflow/infinity"],"tags_count":40,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/infiniflow%2Finfinity","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/infiniflow%2Finfinity/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/infiniflow%2Finfinity/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/infiniflow%2Finfinity/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/infiniflow","download_url":"https://codeload.github.com/infiniflow/infinity/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253745196,"owners_count":21957319,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-native","approximate-nearest-neighbor-search","bm25","cpp20","cpp20-modules","embedding","full-text-search","hnsw","hybrid-search","information-retrival","nearest-neighbor-search","rag","search-engine","tensor-database","vector","vector-database","vector-search","vectordatabase"],"created_at":"2024-07-31T19:00:24.704Z","updated_at":"2026-04-02T12:01:53.756Z","avatar_url":"https://github.com/infiniflow.png","language":"C++","readme":"\u003cdiv align=\"center\"\u003e\n  \u003cimg width=\"187\" src=\"https://github.com/infiniflow/infinity/assets/7248/015e1f02-1f7f-4b09-a0c2-9d261cd4858b\" alt=\"Infinity logo\"/\u003e\n\u003c/div\u003e\n\n\n\u003cp align=\"center\"\u003e\n    \u003cb\u003eThe AI-native database built for LLM applications, providing incredibly fast hybrid search of dense embedding, sparse embedding, tensor and full-text\u003c/b\u003e\n\u003c/p\u003e\n\n\u003ch4 align=\"center\"\u003e\n  \u003ca href=\"https://infiniflow.org/docs/dev/category/get-started\"\u003eDocument\u003c/a\u003e |\n  \u003ca href=\"https://infiniflow.org/docs/dev/benchmark\"\u003eBenchmark\u003c/a\u003e |\n  \u003ca href=\"https://twitter.com/infiniflowai\"\u003eTwitter\u003c/a\u003e |\n  \u003ca href=\"https://discord.gg/jEfRUwEYEV\"\u003eDiscord\u003c/a\u003e\n\u003c/h4\u003e\n\n\nInfinity is a cutting-edge AI-native database that provides a wide range of search capabilities for rich data types such as dense vector, sparse vector, tensor, full-text, and structured data. It provides robust support for various LLM applications, including search, recommenders, question-answering, conversational AI, copilot, content generation, and many more **RAG** (Retrieval-augmented Generation) applications.\n\n- [Key Features](#-key-features)\n- [Get Started](#-get-started)\n- [Document](#-document)\n- [Roadmap](#-roadmap)\n- [Community](#-community)\n\n## ⚡️ Performance\n\n\u003cdiv class=\"column\" align=\"middle\"\u003e\n  \u003cimg src=\"https://github.com/user-attachments/assets/c4c98e23-62ac-4d1a-82e5-614bca96fe0a\" alt=\"Infinity performance comparison\"/\u003e\n\u003c/div\u003e\n\n## 🌟 Key Features\n\nInfinity comes with high performance, flexibility, ease-of-use, and many features designed to address the challenges facing the next-generation AI applications:\n\n### 🚀 Incredibly fast\n\n- Achieves 0.1 milliseconds query latency and 15K+ QPS on million-scale vector datasets.\n- Achieves 1 millisecond latency and 12K+ QPS in full-text search on 33M documents.\n\n\u003e See the [Benchmark report](https://infiniflow.org/docs/dev/benchmark) for more information.\n\n### 🔮 Powerful search\n\n- Supports a hybrid search of dense embedding, sparse embedding, tensor, and full text, in addition to filtering.\n- Supports several types of rerankers including RRF, weighted sum and **ColBERT**.\n\n### 🍔 Rich data types\n\nSupports a wide range of data types including strings, numerics, vectors, and more.\n\n### 🎁 Ease-of-use\n\n- Intuitive Python API. See the [Python API](https://infiniflow.org/docs/dev/pysdk_api_reference)\n- A single-binary architecture with no dependencies, making deployment a breeze.\n- Embedded in Python as a module and friendly to AI developers.  \n\n## 🎮 Get Started\n\nThis section provides guidance on deploying the Infinity database using Docker, with the client and server as separate processes. \n\n### Prerequisites\n\n- CPU: x86_64 with AVX2 support.\n- OS:\n  - Linux with glibc 2.17+.\n  - Windows 10+ with WSL/WSL2.\n  - MacOS\n- Python: Python 3.11+.\n\n\n### Install Infinity server\n\n#### Linux x86_64 \u0026 MacOS x86_64\n\n```bash\nsudo mkdir -p /var/infinity \u0026\u0026 sudo chown -R $USER /var/infinity\ndocker pull infiniflow/infinity:nightly\ndocker run -d --name infinity -v /var/infinity/:/var/infinity --ulimit nofile=500000:500000 --network=host infiniflow/infinity:nightly\n```\n#### Windows\n\nIf you are on Windows 10+, you must enable WSL or WSL2 to deploy Infinity using Docker. Suppose you've installed Ubuntu in WSL2:\n\n1. Follow [this](https://learn.microsoft.com/en-us/windows/wsl/systemd) to enable systemd inside WSL2.\n2. Install docker-ce according to the [instructions here](https://docs.docker.com/engine/install/ubuntu).\n3. If you have installed Docker Desktop version 4.29+ for Windows: **Settings** **\u003e** **Features in development**, then select **Enable host networking**.\n4. Pull the Docker image and start Infinity: \n\n   ```bash\n   sudo mkdir -p /var/infinity \u0026\u0026 sudo chown -R $USER /var/infinity\n   docker pull infiniflow/infinity:nightly\n   docker run -d --name infinity -v /var/infinity/:/var/infinity --ulimit nofile=500000:500000 --network=host infiniflow/infinity:nightly\n   ```\n\n### Install Infinity client\n\n```\npip install infinity-sdk==0.7.0.dev5\n```\n\n### Run a vector search\n\n```python\nimport infinity\n\ninfinity_obj = infinity.connect(infinity.NetworkAddress(\"\u003cSERVER_IP_ADDRESS\u003e\", 23817)) \ndb_object = infinity_object.get_database(\"default_db\")\ntable_object = db_object.create_table(\"my_table\", {\"num\": {\"type\": \"integer\"}, \"body\": {\"type\": \"varchar\"}, \"vec\": {\"type\": \"vector, 4, float\"}})\ntable_object.insert([{\"num\": 1, \"body\": \"unnecessary and harmful\", \"vec\": [1.0, 1.2, 0.8, 0.9]}])\ntable_object.insert([{\"num\": 2, \"body\": \"Office for Harmful Blooms\", \"vec\": [4.0, 4.2, 4.3, 4.5]}])\nres = table_object.output([\"*\"])\n                  .match_dense(\"vec\", [3.0, 2.8, 2.7, 3.1], \"float\", \"ip\", 2)\n                  .to_pl()\nprint(res)\n```\n\n## 🔧 Deploy Infinity using binary\n\nIf you wish to deploy Infinity using binary with the server and client as separate processes, see the [Deploy infinity using binary](https://infiniflow.org/docs/dev/deploy_infinity_server) guide.\n\n## 🔧 Build from Source\n\nSee the [Build from Source](https://infiniflow.org/docs/dev/build_from_source) guide.\n\n## 📚 Document\n\n- [Quickstart](https://infiniflow.org/docs/dev/)\n- [Python API](https://infiniflow.org/docs/dev/pysdk_api_reference)\n- [HTTP API](https://infiniflow.org/docs/dev/http_api_reference)\n- [References](https://infiniflow.org/docs/dev/category/references)\n- [FAQ](https://infiniflow.org/docs/dev/FAQ)\n\n## 📜 Roadmap\n\nSee the [Infinity Roadmap 2025](https://github.com/infiniflow/infinity/issues/2393)\n\n## 🙌 Community\n\n- [Discord](https://discord.gg/jEfRUwEYEV)\n- [Twitter](https://twitter.com/infiniflowai)\n- [GitHub Discussions](https://github.com/infiniflow/infinity/discussions)\n\n","funding_links":[],"categories":["Database","Vector Database Engines","C++","Multidimensional data / Vectors","向量数据库、向量搜索、最近邻搜索","Search","推理 Inference","vector-database","\u003ca name=\"cpp\"\u003e\u003c/a\u003eC++","Awesome Vector Search Engine","Tools"],"sub_categories":["网络服务_其他","Vector search","Standalone Service","General-Purpose Machine Learning"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finfiniflow%2Finfinity","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Finfiniflow%2Finfinity","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finfiniflow%2Finfinity/lists"}