It's odd that the model doesn't support it directly, but they at least have https://docs.z.ai/devpack/mcp/vision-mcp-server
replyOpenrouter definitely supports vision models. Why would you have to give up vision?
reply> Why would you have to give up vision?
replyBecause you would have to switch model.
You can't just say "Oh, button X looks weird see [screenshot]" while coding with GLM. You would need to switch to another model and then maybe back.
For example if I want to paste a screenshot of what I mean, I can't.
replyIf you using opencode or similar you can just temporarily switch models -- in the same session -- to something that has vision and have it look at your image. And then switch back.
replyOr create an agent or subagent that just looks at images, and specify a vision model for that agent.
replyI don't see how that helps, I would still need to somehow get the image into the coding model's context.
replyvision runs just fine locally for most usecases, so it's really just a skill to call that Ollama instance
replyWhy's that?
reply