Add Minimax and Xiaomi as providers
Both can be added through the other option with manual model entry, but it is not intuitive. Anything beside the standard models available through those providers doesn’t work either (e.g. the image-01 and speech-2.8-hd from Minimax).
Both offer very attractive monthly plans and are very capable, would be a great addition!
thank you
Log in to comment and vote
Comments6
Apr 18
PinnedAdded in v2.10
Daniel Nguyen
Apr 13
Wow. Minimax is very powerful. I kinda know about them but didn’t expect their platform to be so good. What kind of tasks do you usually do? I’m adding support for image generation, and TTS, soon will support video generation too.
atomic004
Apr 14
I use it mainly on hermes-agent and before on openclaw, as their token plan is very generous for the extreme token usage those agents have.
In BoltAI2 I use it mainly for smaller projects to help me study, and a whole host of workflows. It is not Opus, but very capable and not bankrupting me.
Daniel Nguyen
Apr 14
Interesting. Then adding support for Minimax in BoltAI wouldn’t help you, no? BTW I’m adding support for Hermes Agent directly, will release soon:
https://help.boltai.com/articles/9536908-how-to-use-hermes-agent-in-boltai
Maybe this is what you actually wanted?
Daniel Nguyen
Apr 13
Thanks. I wanted to understand your request more. Are you asking for API support (using API keys from those platforms), or subscription support? I can see that minimax for example, does offer Token Plan as a subscription. But platforms usually don’t allow consuming this subscription outside of their first party harness.
atomic004
Apr 13
I have the most experience with MiniMax, both their PAYG API and token plan work with an API key (not oauth), and they have guides for integrating with almost any AI app or tool:
https://platform.minimax.io/docs/guides/text-ai-coding-tools
My token plan does work in BoltAI2, but setup is not intuitive, and I had to manually enter the model:
They also support TTS, image, video, music, and lyric generation through their API.
It works fine for me with a small script following this guide, but I would love to use it in BoltAI2:
https://platform.minimax.io/docs/guides/image-generation
As far as I know (though I have less firsthand experience), Xiaomi Mimo and Z.AI GLM are similar:
https://platform.xiaomimimo.com/#/docs/integration/tools-overview
https://docs.z.ai/devpack/overview
Let me know if I can test anything, especially for MiniMax integration.
Thank you for a great product—I really enjoy it.