https://openrouter.ai/api/v1, so there is no endpoint field.
Before you start
Create an account at openrouter.ai, add credit, and generate an API key under Keys.Required Fields
API Key *
Your OpenRouter API key. Keys begin withsk-or-.
Requests fail once your OpenRouter balance reaches zero.
Model Name *
The model identifier in OpenRouter’sprovider/model form, for example anthropic/claude-sonnet-5.5, openai/gpt-6-luna or google/gemini-3.8-flash. Use the exact identifier from the OpenRouter models page, including the provider prefix.
Optional Fields
Model Friendly Name
A label shown in the PipesHub interface so you can tell several configurations apart. If you leave it blank, the model name is used.Context Length
The size of the model’s context window, in tokens. Set it to the value shown for the model on OpenRouter. Leave it empty and PipesHub assumes 128,000.Multimodal
Turn this on if the model accepts images as well as text. It is on by default. Check the model’s page on OpenRouter before you leave it on; if the model refuses images, the check run when you save the model fails.Reasoning
Turn this on if the model reasons before it answers. It is on by default.Configuration Steps
- Open Workspace settings → AI models and click the For LLMs tab.
- Click Configure on the OpenRouter provider card.
- Enter your API Key and the Model Name.
- (Optional) Set the optional fields above.
- Click Add model. PipesHub runs a health check before it saves the model.


Forward user identity
OpenRouter is one of the gateway providers that can receive who is asking with each request. See Identity forwarding.Troubleshooting
- The key is rejected. Check that the key is correct, has not been revoked, and that your OpenRouter account has credit.
- The model is not found. Check the
provider/modelidentifier on the OpenRouter models page. A name without the provider prefix, such asgpt-6-luna, is not found. - The check fails with a message about images. Turn off Multimodal, or choose a model that reads images.