worldly flicker
Hey team, I am trying t get Open Intrpreter to work with my local network Ollama API. It works with OpenAI's API just fine, but I can't get it to talk to my Ollama API.
The API is accessible from the machine I am running Open Interpreter on, and I can make CURL calls to it fine.
I am trying to run it like this:
interpreter --model ollama/solar --api_base http://ts-009.local:11434
I've tried a bunch of variations on the model, all with the same result.
Any help would be much appreciated.
azure bobcat
interpreter --model ollama/solar
thank you Anton. that is what i am doing, plus the url (my api is on another machine on my land), so i am running:
interpreter --model ollama/solar --api_base http://ts-009.local:11434
it returns an error for missing api key:
litellm.exceptions.APIError: OpenAIException - The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable
my api does not need a key. when i give it a fake key, i get this:
litellm.exceptions.NotFoundError: OpenAIException - 404 page not found
i don't think it's trying to hit my api, and i don't know where to add that config
thank you for your help 🙂
azure bobcat
I see
interpreter --model ... --api_base ... --api_key dummykey
interpreter --model ollama_chat/<model-name>
After reading the docs seems like it is ollama_chat not ollama
worldly flicker
thank you, i really appreciate your help. tried that, but still getting:
litellm.exceptions.NotFoundError: OpenAIException - 404 page not found
notice it says OpenAIException - it is still trying to use the OpenAI api, not my local
