1 / 2
【Overview】 Models.dev is an open-source database of AI model specifications, pricing, and features. It gathers model metadata, provider details, and lab information in one place so developers can see what a model supports before committing to it. Instead of checking scattered pages, you can browse the catalog or pull the same data programmatically. The catalog covers models with attributes such as context, output, reasoning, tool call, structured output, temperature support, weights, price, and release or updated dates. Providers are listed with model counts, npm package, API base URL, and docs links, while canonical models are grouped by the lab that authored them. The product is aimed at developers who need reliable model facts for evaluation, integration, or tooling. It is free to use, and every piece of data is reachable either through a browser or through JSON endpoints and a type-safe SDK. 【Key features】 • Model database: Lists AI models with specifications such as context, output, reasoning, tool call, structured output, temperature support, weights, price and release/updated dates. • Provider directory: Shows providers with model counts, npm package, API base URL and docs links. • Labs grouping: Groups canonical models by authoring lab with descriptions and model/provider counts. • JSON API endpoints: Exposes provider data, provider-agnostic model metadata, or the combined catalog through JSON endpoints. • SDK: Provides a type-safe SDK to query the API from an app and exports the latest snapshot for offline use. • Provider logos: Serves provider and lab logos at /logos/{provider}.svg and /logos/labs/{lab}.svg. 【Use cases】 1. Compare AI model specifications and pricing across providers. 2. Query the model catalog via JSON endpoints for scripts or dashboards. 3. Query the API from an app via the SDK with type safety and an offline snapshot. 【Pros and cons】 Pros: - Developers evaluating multiple models get consistent, provider-agnostic metadata without scraping each vendor's site. - The SDK snapshot option suits apps that need to work offline or avoid live API calls for lookups. - Grouping by lab and by provider makes it easy to trace a model back to its author and see everyone serving it. Cons: - JSON API endpoints omit specialized model types by default; you must pass type=all to include all model types. 【FAQ】 Q: What kinds of details does Models.dev record for each model? A: It includes specifications such as context, output, reasoning, tool call, structured output, temperature support, weights, price, and release or updated dates. Q: Can I use the data in my own application? A: Yes. You can query the JSON API endpoints or use the type-safe SDK, which also exports the latest snapshot for offline use. Q: How are models organized on the site? A: Models are shown as provider-agnostic metadata, with model pages listing the providers that serve them and lab pages grouping canonical models by authoring lab.
free
WPS AI is an AI assistant built into WPS Office that helps users write documents, generate presentation outlines and…
Baidu Comate is an AI coding assistant built by Baidu on its ERNIE model. It goes beyond simple autocomplete: it…
UniConverter is an AI media conversion and enhancement tool for video, audio, and images. It combines everyday…
Added
10/20/2026
Views
819
Clicks
5
6 similar tools to consider instead of Models.dev