The project is based on the free policy of Google Gemini API, which provides developers with a daily quota of 1,000 unpaid calls. The quota is applicable to: 1) core scenario testing in the functional verification stage; 2) daily operational needs of small applications; and 3) machine learning teaching in educational institutions. Compared to the paid model of using OpenAI API directly, it can save more than 90% of initial cost.
Economical in that 1) free credits cover the typical MVP development cycle; 2) the excess is still subject to standard Gemini API pricing; and 3) local deployments are exempt from cloud provider data transfer fees. Actual calculations show that small and medium-sized projects can run completely on free credits for the first 3 months.
This answer comes from the articleGemini-CLI-2-API: Converting the Gemini CLI to an OpenAI-compatible Native API ServiceThe
English 
简体中文 
日本語 
Deutsch 
Português do Brasil