stable diffusion ui discord bot
built for use with only one image at a time
Windows:
set "BOTTOKEN=YOUR_BOT_TOKEN"
go run .Linux:
BOTTOKEN=YOUR_BOT_TOKEN go run .Note that you can specify config values from both config.json and enviroment variables. The enviroment variable names are the same as the config.json names except in uppercase.
Default config.json:
{
"bottoken": "",
"channelids": [],
"imagedumpchannelid": "0",
"prefix": "sd!",
"allowbots": false,
"stablediffusionurl": "http://localhost:9500",
"basicauth": "",
"streamimageprogress": 5,
"frameurl": "",
"framehttpbind": ":8080",
"loadingframeurl": "https://c.tenor.com/RVvnVPK-6dcAAAAC/reload-cat.gif",
"defaultprompt": "cat",
"defaultnegativeprompt": "nsfw",
"defaultwidth": 768,
"defaultheight": 768,
"defaultpromptstrength": 0.8,
"defaultinferencesteps": 28,
"defaultguidancescale": 12,
"defaultupscaler": "",
"defaultupscaleamount": 2,
"denychanging": [],
"userslist": {
"whitelistmode": false,
"list": []
},
"chatenabled": false,
"chaturl": "http://localhost:5000/api/latest/generate",
"chatapimode": "kobold",
"chatauth": ""
}Chat mode supports: kobold (http://localhost:5000/api/latest/generate), koboldhorde (https://koboldai.net/api), together (https://api.together.xyz/api/inference), openai (https://api.openai.com/v1/completions), or fallback to simple (http://localhost:8000/generate?input=)
chatauth is basic auth EXCEPT for when openai (it is your openai api key) or koboldhorde (it is your kobold horde token)