Hello everyone! I’ve been trying out the new Claude-code challenge, which I was exited for, I tried to run the code locally but seems that the Openrouter service is paid and unfortunately can’t afford it right now. Is there a way to use another LLM service with a free tier ?? (at least locally, so …

Most LLM providers (local or hosted) should support an OpenAI-compatible API, so I think small changes to your code can get it working against any other provider. Haven’t tried this myself yet though!

thanks for you reply! yeah, I ended up running ollama locally using docker and a gemma3 model

@tomasdepi nice! What changes did you have to make? I assume just pointing the base url to localhost + changing the model name worked?

here are the commands: Run ollama locally docker run -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama Pull the model docker exec -it ollama ollama pull gemma3 btw, gemma3 does not support Tools, which is required for next steps in the challenge, so I’m exploring other mode…

Claude Code API Token

General

Topic		Replies	Views
New challenge: Build your own Claude Code Announcements challenge:claude-code	0	1159	February 4, 2026
Completed the Claude-Code challenge General	2	52	March 8, 2026
Error code 134 in Typescript Bug Reports challenge:claude-code	2	49	February 12, 2026
Language Support: Build your own Claude Code Announcements	0	78	February 23, 2026
Allow commenting on code examples Feature Requests	2	144	June 9, 2024

Claude Code API Token

Related topics