> ## Documentation Index
> Fetch the complete documentation index at: https://ngrok.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Providers

> Built-in and custom AI providers in the ngrok AI Gateway.

Providers are the AI services that host and serve models. The AI Gateway routes your requests to providers and handles authentication, failover, and observability.

Manage providers at [app.ngrok.ai](https://app.ngrok.ai). If you're deciding which setup path to use, start with [Choose how to reach providers](/ai-gateway/guides/configuring-providers). The dashboard separates **built-in** catalog providers from **custom** providers you define.

## Built-in providers

Built-in providers are public AI APIs with a known model catalog: OpenAI, Anthropic, Google, Groq, and others listed in the [model catalog](/ai-gateway/reference/model-catalog). You can send requests with an [access key](/ai-gateway/concepts/access-keys). Your access key authenticates you to the gateway, not to the upstream provider.

For **OpenAI** and **Anthropic**, the AI Gateway can supply upstream credentials when you have [credits](/ai-gateway/concepts/credits). You aren't required to have your own provider account or provider API keys—the AI Gateway handles upstream authentication and bills model cost through credits.

For other built-in providers, [bring your own provider key](/ai-gateway/guides/attaching-provider-keys#add-a-provider-key). Upstream costs go to your provider account; credits still cover the gateway processing fee. This is generally referred to as BYOK (bring your own key).

<CardGroup cols={2}>
  <Card title="OpenAI" icon="robot" href="/ai-gateway/providers/openai">
    GPT and o-series models. ngrok.ai inference available.
  </Card>

  <Card title="Anthropic" icon="a" href="/ai-gateway/providers/anthropic">
    Claude models. ngrok.ai inference available. Supports both OpenAI and Anthropic SDK formats.
  </Card>

  <Card title="OpenRouter" icon="shuffle" href="/ai-gateway/providers/openrouter">
    Access hundreds of models from multiple providers through a single API.
  </Card>

  <Card title="Google" icon="google" href="/ai-gateway/providers/google">
    Gemini models from Google AI Studio.
  </Card>

  <Card title="Groq" icon="bolt" href="/ai-gateway/providers/groq">
    LPU-accelerated inference for open-source models (Llama, Mixtral).
  </Card>

  <Card title="DeepSeek" icon="brain" href="/ai-gateway/providers/deepseek">
    High-performance reasoning and chat models.
  </Card>

  <Card title="Hyperbolic" icon="server" href="/ai-gateway/providers/hyperbolic">
    Open-source model hosting with high-performance inference.
  </Card>

  <Card title="InceptionLabs" icon="sparkles" href="/ai-gateway/providers/inceptionlabs">
    Diffusion-based language models for fast text generation.
  </Card>

  <Card title="Inference.net" icon="network-wired" href="/ai-gateway/providers/inference-net">
    Distributed inference network for AI models at scale.
  </Card>
</CardGroup>

## Custom providers

Custom providers are upstream endpoints you define, such as self-hosted Ollama or vLLM, a private deployment, or any OpenAI- or Anthropic-compatible API that isn't in the built-in catalog. See [Use a model you run yourself](/ai-gateway/guides/use-a-model-you-run-yourself).

Custom providers require a provider key when the upstream needs authentication. See [Custom providers](/ai-gateway/concepts/custom-providers) for the concept.

<CardGroup cols={2}>
  <Card title="Ollama" icon="terminal" href="/ai-gateway/custom-providers/ollama">
    Run open-source models locally with Ollama.
  </Card>

  <Card title="vLLM" icon="bolt" href="/ai-gateway/custom-providers/vllm">
    High-performance inference server.
  </Card>

  <Card title="LM Studio" icon="desktop" href="/ai-gateway/custom-providers/lm-studio">
    Desktop app for local model inference.
  </Card>

  <Card title="Azure OpenAI" icon="microsoft" href="/ai-gateway/custom-providers/azure-openai">
    Microsoft's OpenAI service on Azure.
  </Card>
</CardGroup>

## How provider selection works

When a request arrives, the gateway determines which provider to use:

1. **Explicit provider prefix**: if the model name includes a provider prefix (for example, `openai:gpt-4o` or `openrouter:anthropic/claude-3.5-sonnet`), that provider is used
2. **Catalog lookup**: the gateway looks up the model ID in its [catalog](/ai-gateway/reference/model-catalog) to find the default provider
3. **Request selection**: [model names](/ai-gateway/guides/model-selection-strategies) and fallback lists can override the default

## Next steps

* [Model Catalog](/ai-gateway/reference/model-catalog): browse all available models by provider
* [Bring Your Own Keys](/ai-gateway/concepts/bring-your-own-keys): add provider keys for built-in and custom providers
* [Multi-provider failover](/ai-gateway/examples/multi-provider-failover): automatic failover across providers