Description
ModelRunner offers a comprehensive solution for developers and businesses looking to integrate diverse AI models into their applications without the complexity of managing multiple APIs and backends. It acts as a unified gateway, providing access to hundreds of production-ready models from leading providers such as Google, OpenAI, ByteDance, Stability AI, and Black Forest Labs, all through a single API key.
Read More
The platform supports a wide array of AI capabilities, including image editing, image-to-video, audio and music generation, video editing and enhancement, and text generation. Developers can easily create custom API endpoints (wrappers) for any model, packaging it with preset prompts and selected inputs, which then provides a unique endpoint to call from their applications. This eliminates the need to build or maintain a separate backend infrastructure, as ModelRunner handles output hosting and asset library management automatically.
ModelRunner operates on a transparent, pay-per-use pricing model, where users only pay for what they run, based on each model's published rate (e.g., per image, per second of video, or per megapixel). There are no subscriptions or hidden fees, and pricing is snapshotted at the time of request. This approach allows for cost-effective experimentation and scaling.
For AI assistant developers using the ModelRunner Client Protocol (MCP), integration is streamlined. By pointing an MCP client (like Claude or Cursor) to the ModelRunner MCP server, models become directly callable tools for the assistant, authorized via OAuth without requiring API keys in the configuration. The platform also offers an interactive Playground for testing models with full parameter control before integration, providing example runs and ready-to-use code snippets.
ModelRunner ensures data privacy and security, stating that user inputs and outputs are not used for training. It supports secure authentication methods, including passkeys, OAuth (GitHub, Google), and two-factor authentication. Users have control over data retention, with options to shorten retention, set account-wide defaults, opt out of storing payloads, or delete request data and generated files at any time. The platform also supports enterprise workloads with high throughput and reliability through its async queue system, offering dedicated capacity, custom SLAs, and volume pricing for larger needs. Additionally, ModelRunner provides free API credits for active open-source projects through its Open Source Program.
- CategoryCoding & Development
- Pricing ModelPay-per-use
- DeploymentCloud
- PlatformsWeb
ModelRunner's Core Features
Unified API for AI models
Access to image, video, audio, 3D, and text models
Integration with providers like Google, OpenAI, ByteDance, Stability AI
Single API key for all models
Consistent request shape across models
Custom API endpoint creation (wrappers)
No backend required for integration
Automatic output hosting and asset library
Pay-per-use pricing model
No subscriptions or hidden fees
MCP client integration for AI assistants
Interactive model Playground for testing
Secure authentication and data privacy controls
Support for enterprise workloads
Open Source Program with free credits
How to use ModelRunner?
Create Account: Sign up for a ModelRunner account.
Add Credits: Purchase credits to fund your usage.
Generate API Key: Create an API key for authentication.
Explore Models: Browse the catalog and test models in the Playground.
Integrate: Make requests via the REST API or JavaScript SDK.
Deploy Wrappers: Package models into custom API endpoints for your app.
ModelRunner's Use Cases
- AI Feature Development
- Multi-Model Integration
- AI Assistant Tooling
- Custom AI Endpoints
- Cost-Effective AI Usage
- Rapid Prototyping
- Enterprise AI Solutions
- Open-Source Project Support
