For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Help us improve OpenRouter by reporting issues with AI generations. You can submit feedback directly from the Chatroom or the Activity page.
Overview
The Report Feedback feature allows you to flag problematic generations with a category and description. This helps our team identify and address issues with model responses, latency, billing, and more.
Feedback Categories
When reporting feedback, select the category that best describes the issue:
Latency: Response was slower than expected
Incoherence: Response didn’t make sense or was off-topic
Incorrect Response: Response contained factual errors or wrong information
Formatting: Response had formatting issues (markdown, code blocks, etc.)
Billing: Unexpected charges or token counts
API Error: Technical errors or failed requests
Other: Any other issue not covered above
Reporting from the Chatroom
In the Chatroom, you can report feedback on individual assistant messages:
Hover over an assistant message to reveal the action buttons
Click the bug icon to open the Report Feedback dialog
Select a category that describes the issue
Add a comment explaining what went wrong
Click Submit to send your feedback
The generation ID is automatically captured from the message, so you don’t need to look it up.
Reporting from the Activity Page
The Activity page offers two ways to report feedback:
Per-Generation Feedback
Each row in your activity history has a feedback button:
Click the Report Feedback button in the header (top right)
Enter the generation ID (found in your API response or activity row)
Select a category and add your comment
Click Submit
Finding Your Generation ID
The generation ID is returned in the API response under the id field. You can also find it by clicking on a row in the Activity page to view the generation details.
What Happens After You Submit
Your feedback is reviewed by our team to help improve:
Model routing and provider selection
Error handling and recovery
Billing accuracy
Overall platform reliability
We appreciate your help in making OpenRouter better for everyone.