This topic was automatically generated from Slack. You can find the original thread here.
Hello team,
I am from Runware (https://runware.ai/). We’ve been notified of a bug related to the Pipedream-Runware MCP integration.
The issue is that while generating images using the new Gemini Flash Image 2.5 model. It appears the MCP server is including seed and dimension parameters in the API request, which the model doesn’t support. These params were made optional. Can someone help us with this or can you give us access to update it?