The app generates random images using AI Stable Diffusion and posts them in Telegram https://t.me/ai_art_random
Environment variables:
HUGGING_FACE_HUB_TOKEN
- token for Hugging Face for downloading modelsTELEGRAM_TOKEN
- bot tokenTELEGRAM_ADMIN_ID
- user ID to manage the botTELEGRAM_CHAT_ID
- chat where images will be sentUPSCALING
- up to 4x image resolution with Real-ESRGAN (defaulttrue
)COMMAND_ONLY_MODE
- bot command mode onlyFACE_RESTORING
- restoring faces in an image with GFPGAN (required enabled upscaling)SLEEP_TIME
- how many seconds to sleep between generations (default 60s)
Build and running in docker:
docker build -t ai-art-bot .
docker volume create ai-art-bot_huggingface
docker run --name ai-art-bot -d --env-file .env -v ai-art-bot_huggingface:/root/.cache/huggingface --gpus=all ai-art-bot
Running with docker-compose:
docker-compose up -d