Skip to content

Models.dev

An open-source database for AI model specifications, pricing, and capabilities.

Freemium · Web · API

stork0:00 / 0:04
Ava, one of 89 voices. Press play.
A voiceover costs 5 cents a minute.
Type up to 20,000 characters, get an MP3. 89 voices, 38 languages. Three free a day, a minute each.
Make a voiceover

What is Models.dev?

Models.dev is an AI model metadata database tool developed by anomalyco that enables API builders, AI product teams, and platform engineers to access comprehensive specifications, pricing, and features of various AI models. It serves as a centralized, structured source of truth for AI model metadata, addressing the challenge of scattered information across different provider documentation. The platform functions as a comprehensive, open-source database of AI model specifications, pricing, and features, rather than a model router or an inference provider. It offers a structured resource for quickly gathering facts about AI models before committing to testing or integration. Models.dev is actively maintained, with recent discussions (June-July 2026) highlighting its utility, particularly in conjunction with tools like OpenCode, which leverages Models.dev internally for model selection. The project's GitHub repository shows ongoing contributions, with a merge pull request from August 2025, indicating continuous data updates. User reception is generally positive, with developers appreciating its open-source nature and its ability to simplify the initial screening process for AI models, though some feedback suggests a desire for improved user interface/experience.

Key Features of Models.dev

Models.dev provides a robust set of features designed to centralize and standardize information about AI models for developers and product teams.

  • Access to a wide range of AI models from various providers.
  • Comprehensive open-source database of AI model specifications.
  • Detailed pricing information for cataloged AI models, including token costs.
  • Capabilities information for AI models, such as context limits and modalities.
  • API available for programmatic access to AI model data via JSON endpoints.
  • Structured metadata for model-agnostic agents and coding tools.
  • Comparison functionalities for AI model pricing, context, modalities, and capabilities.
  • Community-contributed data ensuring broad coverage and ongoing updates.

Who Should Use Models.dev?

Models.dev is primarily designed for technical professionals and teams involved in AI product development and integration, seeking a reliable source for AI model information.

  • API Builders: To access AI model data programmatically via an API for internal tools and comparisons.
  • AI Product Teams: For checking basic facts about AI models before integration into products and building a shortlist for further evaluation.
  • Platform Engineers: To compare AI model pricing, context, modalities, and capabilities for system design.
  • Developers: For providing structured metadata for model-agnostic agents and coding tools, especially for agentic AI development requiring specific features like tool calling support.

How to Use Models.dev

Utilizing Models.dev involves accessing its web interface or programmatic API to retrieve detailed information about various AI models.

  1. Access the Models.dev website at https://models.dev/ to browse the model catalog.
  2. Utilize the search and filter functions on the platform to identify specific model specifications, pricing, and features.
  3. Consult the API documentation at https://github.com/anomalyco/models.dev#api to programmatically access AI model data.
  4. Integrate the Models.dev API into internal tools or applications for automated model selection and data retrieval.
  5. Contribute to the open-source database by submitting updates or new model information via its GitHub repository.

Models.dev Pricing & Plans

Models.dev operates as an open-source database, meaning there are no direct costs or pricing plans associated with using the Models.dev tool or accessing its data. The platform itself is free to use and access. Its primary function, however, is to provide comprehensive pricing details for the various AI models it catalogs from different providers. These pricing details are specific to the individual model providers and can vary significantly, often including input and output token costs per million tokens, and sometimes cache-read and cache-write costs. Therefore, while Models.dev is free, the AI models it describes may incur usage-based costs from their respective providers.

  • Models.dev Platform: Free (open-source database access).
  • Cataloged AI Models: Pricing varies by individual model provider (e.g., per-token costs for LLMs).

Pros

  • Provides a centralized, open-source database for comprehensive AI model specifications, pricing, and capabilities.
  • Offers programmatic access via a JSON API, facilitating integration into internal tools and automated workflows.
  • Enables detailed comparative analysis of AI models across various providers based on objective facts.
  • Supports initial model shortlisting for API builders and AI product teams, streamlining evaluation processes.
  • Actively maintained and community-contributed, ensuring ongoing updates to model data.
  • Useful for agentic AI development by providing critical features like tool calling support and context window sizes.

Cons

  • Models.dev is a metadata provider, not an inference platform; it does not host or run AI models.
  • Relies on community contributions for data accuracy and timeliness, which can vary.
  • User interface/experience has been noted by some users as an area for potential improvement.
  • The platform's status feed is officially linked to https://status.openai.com, which may cause confusion regarding its own operational status.
  • Recorded 83 incidents in the last 90 days, with the last incident on 2026-07-01, indicating potential service reliability concerns for its own operations.

Models.dev vs Competitors

Models.dev differentiates itself by focusing on being an open-source, community-contributed *metadata database* for AI models, rather than an inference provider or a unified API gateway. Its core value lies in consolidating scattered information into a single, queryable dataset.

Models.dev

Freemium · Web · API

From the Stork blog

Enjoying this? Get one like it in your inbox each morning.

one email a day · unsubscribe in two clicks · no third-party tracking