One API for the models your product needs

Pick a model. Call it. Switch any time.

Model APIs on OneInfer Cloud put text, vision, image, audio and video models behind one OneInfer API key. Compare models, providers and input and output prices in the catalog, then call chat models with the OpenAI SDK you already use: change the base URL and the key, and keep the rest of your code.

Test requests in the playground and follow usage and credits in the Console. When a workload needs capacity of its own, move it to a dedicated endpoint on the same account.

Browse modelsGet an API key

Your first call in 4 steps

Step 1 of 4

Create a key in the Console under API keys. One key works across the models in the catalog.

export ONEINFER_API_KEY="your-api-key"

Start with three practical steps

OneInfer removes the provider-by-provider setup loop so your team can evaluate models and begin integrating from one place.

  1. Choose a model

    Compare supported models, providers, context windows, and input/output pricing in one catalog.

    Browse model catalog

  2. Create one API key

    Create and manage your OneInfer key from the console. The same credential works across the unified API surface.

    Create API key

  3. Send your first request

    Use a familiar OpenAI-compatible request shape, select a model ID, and start building without provider-specific setup.

    Open playground

Change the endpoint. Keep building.

Use an OpenAI-compatible chat request for supported language models. Your OneInfer key handles access while the model ID selects the workload you want to run.

  • One credential managed from the console
  • Model IDs discoverable in the public catalog
  • Usage and credits visible in one account
Read the API documentation

OneInfer request path

  • Your application

    One key + model ID

  • OneInfer control plane

    Auth, metering, routing

  • Model execution

    Provider + runtime

API surface
Unified
Billing
Usage based
Visibility
Console

More than an endpoint proxy

The API sits inside the wider OneInfer control plane, connecting model discovery, access, execution, observability, billing, and the path to dedicated infrastructure.

  • One API across modalities

    Use a consistent OneInfer surface for language, vision, image, audio, and video workloads.

  • Model and provider freedom

    Move between frontier, open, and specialist models without rebuilding authentication and billing for every provider.

  • Transparent unit economics

    See model input and output rates before integration, then monitor credit usage from the console.

  • Production observability

    Track requests, usage, model behavior, and spend from the same control plane used to manage access.

  • Intelligent execution paths

    Use OneInfer routing to balance model fit, provider availability, latency, and cost for supported workloads.

  • A path beyond shared APIs

    Move demanding workloads to intelligent endpoints or dedicated deployments without leaving the platform.

Find the right model before you integrate

View every model

Start shared. Add control when you need it.

  1. Start: Model APIs

    Explore models and ship your first integration using shared, usage-based access.

    Explore models

  2. Optimize: Intelligent Endpoints

    Control routing and combine endpoints when traffic, reliability, or model selection becomes more complex.

    Open console

  3. Scale: Dedicated Deployments

    Provision dedicated model capacity when your workload needs isolated resources and predictable control.

    View deployments

Pick a model. Create a key. Make the call.

Start in the catalog, test your request in the playground, and use the same OneInfer account as your application grows.

Explore model APIsOpen playground