Hi, I'm having issues using minimax via openrouter...
# openhands-cloud
l
Hi, I'm having issues using minimax via openrouter for some reason, I've added the openrouter base url, api key and model in the advanced section of the LLM page but I'm having issues still
b
Are you specifically trying to use the openrouter one? Why not use it through the OpenHands provider which I believe is free.
If doing it through OpenRouter, can you show how you configured the settings.
l
that'd be a temporary fix,we need to keep using openrouter to keep our llm usage in a single place
Screenshot 2026-02-24 114816.png
here you go @better-park-34863
litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed model=minimax/MiniMax-M2.5 Pass model as E.g. For 'Huggingface' inference endpoints pass in
completion(model='huggingface/starcoder',..)
Learn more: https://docs.litellm.ai/docs/providers Here's the error
b
In the custom model try: openrouter/minimax/MiniMax-M2.5
l
new error: litellm.BadRequestError: OpenrouterException - {"error":{"message":"Budget has been exceeded! User=6395c53d-f921-440d-9c2c-d05026df087e in Team=6395c53d-f921-440d-9c2c-d05026df087e Current cost: 0.0, Max budget: 0.0","type":"budget_exceeded","param":null,"code":"400"}}
b
That seems a bit better? You have credits on the account I'm guessing.
l
I do
yea, it seems like progress though lol
b
Try removing the base URL. Do you get a different error or same error.
l
litellm.AuthenticationError: AuthenticationError: OpenrouterException - {"error":{"message":"Authentication Error, Invalid proxy server token passed. Received API Key = sk-...84 . . Unable to find token in cache or `LiteLLM_VerificationTokenTable`","type":"token_not_found_in_db","param":"key","code"
I can see that it's now https://llm-proxy.app.all-hands.dev automatically, so probably routing via all-hands proxy url but the api key is of openrouter which is a mismatch
b
Huh yeah okay so put the base url back that seemed to hit the right one at least.
I'm looking at the openrouter docs on litellm: https://docs.litellm.ai/docs/providers/openrouter
Base URL seems to be: https://openrouter.ai/api/v1 Not sure if that will make a difference
l
yep, let me try
b
And then the model will be: openrouter/minimax/minimax-m2.5
l
yea, the error seems to be about budget or something
b
So it's likely we are doing something on our side.
@limited-student-10747 sorry to ping you directly, but any idea if we are messing up users who want to use minimax with openrouter?
l
litellm.AuthenticationError: AuthenticationError: OpenrouterException - {"error":{"message":"Authentication Error, Invalid proxy server token passed. Received API Key = sk-...8421, Key Hash (Token) =5b. Unable to find token in cache or `LiteLLM_VerificationTokenTable`","type":"token_not
now I am getting this error for the initial base url as well
hey, started a fresh convo and it seems to be working!
b
Oh nice!
l
thanks a lot for the help man!
np 1
🎉 1
I was wondering what are you guys using under the hood to spin up these sandboxes? daytona or something?
b
No. It's in our own cloud infrastructure
💯 2