- support wgcf to unblock access denied
- support all
backen-api
- support login and auto get
ACCESS_TOKEN
- support manually get
ACCESS_TOKEN
- support use
ACCESS_TOKEN
- support proxy
- base on browser
- Auto Click cloudflare checkbox
- auto refersh access_token when 403
- support arm64,amd64
- vnc password
- auto login
- network access to openai
- openai account
- 500M Memory, you can enable swap on you server
Options:
- set
auto_refersh_access_token=True
. Openlocalhost:7900
, your can forward port 7900 to local by ssh, and login tohttps://chat.openai.com/
, if you get 403 when run sometimes. pleaselocalhost:7900
and check again. Make sure the first tab of the browser ishttps://chat.openai.com/
- set
ACCESS_TOKEN
in your requset header'sAuthorization
ssh -L 7900:localhost:7900 azure_vm
UI use chatgpt-web-share
change config.yaml
# from
chatgpt_base_url: http://127.0.0.1:6060/api/
# to
chatgpt_base_url: http://chatgpt-reverse-proxy:8000/backend-api/
version: "3"
services:
chatgpt-share:
image: ghcr.io/moeakwak/chatgpt-web-share:latest
container_name: chatgpt-web-share
restart: always
ports:
- 80:80
- 8080:80
# - 6060:6060
depends_on:
chatgpt-reverse-proxy:
condition: service_healthy
volumes:
- ./data:/data
- ./config.yaml:/app/backend/api/config/config.yaml
- ./logs:/app/logs
chatgpt-reverse-proxy:
image: ghcr.io/dreamhunter2333/chatgpt_reverse_proxy:latest
# if you use arm64
# image: ghcr.io/dreamhunter2333/chatgpt_reverse_proxy:arm64
container_name: chatgpt-reverse-proxy
hostname: chatgpt-reverse-proxy
restart: always
ports:
- "8000:8000"
- "7900:7900"
# if use auto_refersh_access_token please mount tmp
# volumes:
# - ./tmp:/app/tmp
# if you ip is access denied, you can uncomment this to use wgcf
# network_mode: "service:wgcf"
environment:
- VNC_NO_PASSWORD=1
# do not set VNC_NO_PASSWORD and you can set VNC_PASSWORD
# - VNC_PASSWORD=password
- user_data_dir=/app/tmp
# - timeout=100000
# - navigation_timeout=100000
# - proxy=http://127.0.0.1:7890
# - auto_refersh_access_token=False
healthcheck:
test:
[
"CMD-SHELL",
"curl --fail http://localhost:8000/health_check || exit 1"
]
interval: 30s
timeout: 10s
retries: 10
# if you ip is access denied, you can uncomment this to use wgcf
# wgcf:
# image: neilpang/wgcf-docker:latest
# volumes:
# - ./wgcf:/wgcf
# - /lib/modules:/lib/modules
# privileged: true
# sysctls:
# net.ipv6.conf.all.disable_ipv6: 0
# cap_add:
# - NET_ADMIN
# command: "-4"
python3 -m venv ./venv
./venv/bin/python -m pip install -r requirements.txt
./venv/bin/playwright install
./venv/bin/python server.py
./venv/bin/python main.py
build docker image
docker compose buid
docker compose push
docker buildx build --platform linux/amd64 . --file Dockerfile --tag ghcr.io/dreamhunter2333/chatgpt_reverse_proxy:latest --push
docker buildx build --platform linux/arm64 . --file Dockerfile --tag ghcr.io/dreamhunter2333/chatgpt_reverse_proxy:arm64 --push