Repository navigation
Cannot use iflowcn/minimax-m2 model in OpenCode. Error status 435 "Model not support". #4305
Description
Description
I am trying to use iflowcn/minimax-m2, but it doesn't output anything at all. Other coding CLIs, such as iFlow CLI, and other models of the same provider work just fine. I'll provide the logs as txt files; if y'all want another format, just ask me. I also tried checking if the model name is wrong or wth, but it's not it's the same on iFlow CLI.
OpenCode version
1.0.63 via npm
Steps to reproduce
- Open Opencode
- Select Minimax M2
- Send some prompt
Screenshot and/or share link
iflow_minimax-m2.txt
iflow_glm4.6.txt
opencode_glm4.6.txt
opencode_minimax.txt
Operating System
Fedora Linux 43 with distrobox/ Windows 11 via wsl tried both
Terminal
Konsole/ Windows Terminal
Activity
github-actions commentedon Nov 13, 2025
This issue might be a duplicate of existing issues. Please check:
Feel free to ignore if none of these address your specific case.
redzrush101 commentedon Nov 13, 2025
Ah sorry I didn’t notice the closed issues
rekram1-node commentedon Nov 13, 2025
Hm never used the provider, is it defined wrong here at all?
https://github.com/sst/models.dev/blob/dev/providers/iflowcn/models/minimax-m2.toml
Maybe they dont support that model via
@ai-sdk/openai-compatibledo u have to use the anthropic endpoint for it?redzrush101 commentedon Nov 13, 2025
Nah, there's no Anthropic endpoint. Can cache_read = 0.00 and cache_write = 0.00 break the model Does it force caching? The provider doesn’t support caching ig. The rest of the configuration is fine it; I just based it off of OpenRouter I’m jst gonna go and ask them on their forum tomorrow.
rekram1-node commentedon Nov 13, 2025
The caching stuff is just on our end for calculations it shouldn't be sending anything because of that field
redzrush101 commentedon Dec 15, 2025
imma jst close this old issue I asked them in their forum they say they dont provide any support for this model outside of their own client for some reason