About llmdb.xyz
What is this?
llmdb.xyz is a comprehensive database of Large Language Models (LLMs) from all major providers. Browse, filter, and compare models by capabilities, pricing, context windows, and more.
This site is powered by llm_db, an open-source Elixir package that provides a unified interface for querying LLM model metadata across providers.
The llm_db package is designed to power req_llm, a comprehensive Elixir AI API client that provides a unified interface for interacting with OpenAI, Anthropic, Google, Mistral, and many other LLM providers.
Features
- Browse 2,000+ models from OpenAI, Anthropic, Google, Mistral, and more
- Filter by capabilities: chat, embeddings, reasoning, tool use, streaming
- Compare pricing across providers
- View context windows and output limits
- Filter by input/output modalities (text, image, audio)
- Dark mode support
Data Sources
The llm_db package aggregates model metadata from multiple sources to build a comprehensive catalog spanning 127+ providers. Data is collected through a build-time ETL pipeline that ingests, normalizes, validates, and merges records from each source:
- models.dev — community-maintained model catalog with pricing, context limits, and capabilities
- OpenAI Models API — official model list directly from OpenAI
- Anthropic Models API — official model list directly from Anthropic
- Google Gemini API — official model list from Google
- xAI API — official model list from xAI
- OpenRouter — aggregated model catalog across dozens of providers
- Hand-curated local TOML overrides for corrections and additional metadata
Sources are merged with a last-wins precedence strategy, allowing hand-curated data to override upstream records where needed.
Acknowledgements
Special thanks to the models.dev project for maintaining an open, community-driven catalog of LLM model metadata. Their comprehensive dataset serves as a foundational data source for llm_db and makes projects like this one possible.
Open Source
Both this website and the underlying llm_db package are open source.
Author
Built by Mike Hostetler. Found a bug or have a suggestion? Please submit an issue on GitHub.
Community
Join us on Discord to discuss LLMs, share feedback, or contribute to the project.
Join Discord