Integrations
IQ Routing is one gateway in front of a broad set of model providers. Bring your own API key for whichever providers you already use, point your client at a single base URL, and every provider you’ve configured is reachable behind it without juggling separate SDKs or endpoints. You remain each provider’s customer of record, on your own account and your own key.
Supported providers
Add credentials for any of the following from the dashboard. Nothing is enabled until you add a key, and every org’s credentials are encrypted at rest.
- AWS Bedrock
- Anthropic
- Cerebras
- Cohere
- DeepInfra
- DeepSeek
- Fireworks AI
- Groq
- MiniMax
- Mistral
- Moonshot AI (Kimi)
- OpenAI
- OpenRouter
- Qwen (Alibaba Cloud)
- Replicate
- Together AI
- xAI
- Z.ai (GLM)
Custom endpoints
On the Team plan and up, you can also point IQ Routing at your own OpenAI- or Anthropic-compatible endpoint -- your own URL and your own key -- instead of, or alongside, a named provider. The endpoint must be HTTPS and publicly reachable. See the provider docs for setup details.
A note on provider location
Five of the providers above -- DeepSeek, Z.ai (GLM), Moonshot AI (Kimi), Qwen (Alibaba Cloud), and MiniMax -- are headquartered in China. As with every provider on this page, bring-your-own-key means a routed request is authenticated with your own key and served directly by that provider’s own infrastructure. See Subprocessors for more detail.
Next steps
For setup steps, authentication, and per-provider notes, see the provider docs. For the model families and ids available through each provider, see the model list. To start routing traffic, head to the quickstart.
IQ Routing is not affiliated with, endorsed by, or sponsored by any of the providers listed above. All product names and trademarks are the property of their respective owners.