As recorded at collection: Free use of the Gugugaga (咕咕嘎嘎) AI API

Gugugaga (咕咕嘎嘎) is currently a free, open-access site that offers unlimited Grok and a small selection of text models. The text models are deepseek-v4-flash, glm-5.2, the Claude family, and the GPT family. The image generation models are grok-imagine-image-2.0, gpt-image-2, gemini-3-pro-image-preview, and gemini-3.1-flash-image-preview.

Visit api.fengshao1227.com to check the API documentation, then request the desired model using the call format specified in the docs. The site has a LINUX DO Connect integrated login feature. The image generation request path is v1/images/generations. For the GPT image example, set model to gpt-image-2, prompt to 'Convert to a 3D style', size to 1024*1024, quality to high, and image to the image URL. For the Gemini image example, set model to gemini-3-pro-image-preview, prompt to the same description, sizaspectRatioe to 9:16, resolution to 1K, and image to the image URL. The documentation indicates that resolution supports only 1K and 2K.

The publisher says this is a personal project unrelated to any company or commercial organization, with no effectively paid features such as usage fees or donations. The available models are currently limited to some. grok-imagine-image-2.0 cannot edit images via the web relay method, and the Console management interface relay method is currently experiencing repeated errors.