This is a tool that allows anyone to host an Ollama online inference server.
You can use spare PCs to perform inference through Ollama.
It's like a personal version of Colab, which is provided by a certain Google.
For hosts, if you're on Windows, run init_server.bat, otherwise run init_server.sh. If you have everything necessary, you can set up a server right away. Then you can share the URL with anyone who wants to access it, and they can access the chat screen. (It's compatible with OpenAI API. When using it, add /v1 to the end of the URL that appears.)
The server basically does not send information. Additionally, we ensure that chat history and parameter settings such as system prompts are not retained. If you want to reuse them, you can import/export JSON, so please manage it on the user side.
Token usage and other metrics are also recorded in CSV format. This makes it easy to check if you want to charge for token usage or similar. You are permitted to modify this package or create billing flows. In that case, please modify it in accordance with applicable laws and regulations. I assume no responsibility.
Reactions & commentsFeedback
Share your thoughts, bug reports, or suggestions directly with the developer
Log in to share your feedback
Log in to leave feedback