Following on from my openai-chatgpt-terminal project to add ChatGPT functionality to my terminal, I wanted something self hosted on my homelab - and can be installed app-like on my phone so moved this project to a container. Simple webapp utilising python's Flask and OpenAI module to provide a simple chat interface; able to choose between different API models (gpt-4,gpt-3.5-turbo), refresh, dark mode and code block formatting and one click copy to clipboard.
Pull from dockerhub - dockerhub | jamescwylde/openai-chatgpt-docker:
- ARM:
docker pull jamescwylde/openai-chatgpt-docker:arm
- x86:
docker pull jamescwylde/openai-chatgpt-docker:latest
Pass your OpenAI API key as an env variable, -e OPENAI_API_KEY="API_HERE". i.e
docker run -d -p 6565:6565 -e OPENAI_API_KEY="API_HERE" --network vnet jamescwylde/openai-chatgpt-docker:arm
Change, build and tag:
git clone https://github.com/jameswylde/openai-chatgpt-docker-container.git
cd openai-chatgpt-docker-container
docker buildx build --platform linux/amd64 -t yourname/projectname:latest .
docker run -d -p 6565:6565 -e OPENAI_API_KEY="API_KEY" yourname/projectname:latest