With GPT4Free, you can use these large models for free. This project works in a clever way: it disguises itself as the official chat client of chatgpt, or some other websites that have already paid OpenAI, such as poe, bing, etc., and then indirectly calls OpenAI's services by calling the interfaces of these websites.
Moreover, deployment is super simple. With just two Docker commands, you can easily use GPT-4 locally:
docker pull hlohaus789/g4f
docker run -p 8080:8080 -p 1337:1337 -p 7900:7900 --shm-size="2g" hlohaus789/g4f:latest
Open your browser and enter http://localhost:8080 to open the chat page.
The implementation principle of this project is actually very simple, it disguises itself as the official chat client of ChatGPT or some other websites that have already paid OpenAI and can use OpenAI services normally, by calling the interfaces of these websites to indirectly use OpenAI (including Bard, Claude, etc.).
Interested friends can try to deploy it themselves, and with Docker, it can be done in 2 minutes. The author has also put the service on Discord, so you can experience it by entering the chat channel.
For more detailed information, including how to call the API, please check the project's GitHub address: