Infinity
Infinity is an open-source database for LLM applications, designed for retrieval-augmented generation and hybrid search.
What Infinity is
Infinity is an open-source database for LLM applications, designed for retrieval-augmented generation and hybrid search. It searches dense and sparse embeddings, tensors and full text, with filtering across data types such as strings, numbers and vectors. Listed rerankers include RRF, weighted sum and ColBERT. The project homepage claims 0.1 millisecond query latency and up to 15K QPS on million-scale vector datasets. Developers can work through Python and HTTP APIs; the documentation includes a Python SDK and pip installation. Deployment options described in the documentation include Docker and Linux packages in deb, rpm or tgz formats. Supported environments include Linux, macOS and Windows 10 or later with WSL or WSL2, on x86_64 systems with AVX2; Python 3.10 or later is required. Search has configuration considerations: conditional filters must work through an index, and full-text indexing requires a tokenizer or analyzer. The guide says HNSW targets faster dense-vector search, while IVF uses less memory. Streaming vectors are searched by brute force alongside indexed results, and tensor search can have significant storage overhead. Infinity is free and open source under the Apache-2.0 license.
Good
- Supports hybrid search across dense, sparse, tensor and full-text data.
- Provides Python and HTTP APIs.
- Docker and Linux package deployment are documented.
- Free and open source under Apache-2.0.
Watch out
- Conditional filters must work through an index.
- Full-text indexing requires a tokenizer or analyzer.
- Tensor search can add significant storage overhead.
- Client and server versions must match exactly in client-server mode.
Who it suits
Infinity may suit developers building retrieval-augmented generation applications who need hybrid search and filtering. Its documented Python and HTTP APIs and multiple deployment options are aimed at technical users.
Verdict
Infinity offers a broad hybrid-search toolkit and documented deployment routes for LLM applications. Its search setup has specific indexing requirements, and tensor search may increase storage needs.
The facts
Each one read from Infinity's own pages, with the date it was checked.
What it does
- API options
- The docs provide Python and HTTP API references, and the homepage says Infinity has a Python API.checked 8 Oct 2026
- Cluster storage
- The cluster guide says supported shared storage is MinIO and describes a cluster with one leader, up to four followers, and learner nodes.checked 8 Oct 2026
- Community
- The homepage links to Infinity's GitHub repository and Discord community.checked 8 Oct 2026
- Community support
- The homepage links to the project’s X account, GitHub, and Discord community.checked 8 Oct 2026
- Data types
- Infinity supports data types including strings, numerics, and vectors.checked 8 Oct 2026
- Deployment
- The quickstart documents deploying Infinity with Docker, while the binary guide offers Linux packages in deb, rpm, or tgz formats.checked 8 Oct 2026
- Developer interface
- The homepage describes an intuitive Python API, and the docs provide a Python SDK and HTTP API reference.checked 8 Oct 2026
- Full-text search
- The search guide says full-text indexing requires specifying a tokenizer or analyzer, and most built-in analyzers require resource files.checked 8 Oct 2026
- Intended use
- The maker describes Infinity as designed for retrieval-augmented generation applications.checked 8 Oct 2026
- License
- The project’s GitHub repository lists the Apache-2.0 license.checked 8 Oct 2026
- Performance
- The homepage states 0.1 milliseconds query latency and up to 15K QPS on million-scale vector datasets.checked 8 Oct 2026
- Performance claims
- The homepage claims 0.1 millisecond query latency and up to 15K QPS on million-scale vector datasets.checked 8 Oct 2026
- Purpose
- Infinity describes itself as an AI-native database for LLM applications that provides hybrid search.checked 8 Oct 2026
- Python API
- The homepage describes an intuitive Python API, and the docs show installing the client with pip.checked 8 Oct 2026
- Ranking
- Its listed rerankers include RRF, weighted sum, and ColBERT.checked 8 Oct 2026
- Rerankers
- The homepage names RRF, weighted sum, and ColBERT as supported rerankers.checked 8 Oct 2026
- Search
- It supports hybrid search across dense embeddings, sparse embeddings, tensors, and full text, with filtering.checked 8 Oct 2026
- Search constraints
- The search guide says conditional filters must work through an index; secondary-index filters support logical comparisons and membership expressions.checked 8 Oct 2026
- Search methods
- It supports hybrid search across dense embeddings, sparse embeddings, tensors, and full text, as well as filtering.checked 8 Oct 2026
- Supported environments
- The quickstart lists Linux, macOS, and Windows 10 or later with WSL or WSL2; it specifies x86_64 with AVX2 support and Python 3.10 or later.checked 8 Oct 2026
- Tensor guidance
- The search guide cautions that tensor search can have significant storage overhead and recommends tensors for reranking in hybrid search.checked 8 Oct 2026
- Vector search
- The search guide says HNSW is aimed at faster dense-vector search and IVF uses less memory; incoming streaming vectors are searched by brute force alongside indexed results.checked 8 Oct 2026
- Version requirement
- The Python API documentation says client and server versions must match exactly in client-server mode.checked 8 Oct 2026
Alternatives to Infinity
- 1
Chroma
Free planOpen sourceAndroidiOSWebAPI8.6From $250 a month - 2
Vespa
Open sourceWebAPILinuxmacOS8.4 - 3
Pinecone
Free planWebAPILinuxmacOS7.8From $20 a month - 4
Weaviate
Free planWebAPILinuxSelf-hosted7.0$45 usage-based - 5
Qdrant
Free planWebAPILinuxmacOS6.9Free plan - 6
Elasticsearch
WebAPILinuxSelf-hosted6.7From $99 a month