You can modify instruction text for ChatGPT by property in config.json.In the beginning of the question ChatGPT reads, the questioner's username will be attached in format of "*username*さんからの質問:" (EN: Question from *username*: ).If both is specified config.json's value will be used. This number can be changed from MAX_HISTORIES environment variable (it's defined in. As default, last 10 conversation will be saved as context. Message context (a history of conversation) is saved per channel.Copy the Bot Token and save it for later use.Ĭopy.(Please make sure to turn PUBLIC BOT switch OFF if you don't want to reveal your bot to the others.) Enter a name for your application and click "Create".Copy the API key and save it for later use.Click profile icon on right top then click 'View API keys' ( here).Setup instructions for lambda function is currently not ready Getting Startedīefore starting your bot, you need to get OpenAI API key and Discord Bot token. Bot timeouts due to slow startup during cold boot). Lambda function bot releases you from consistently launching the bot process on your local PC.īut this feature is currently in progress so there are some problems(i.e. the "gateway" function to handle defer reply and async call of "main" function.the "main" function to handle chatgpt request/response.Lambda function bot (Unstable)Īs the another option to the daemon bot, you can run this bot on your AWS environment, as two AWS lambda functions: This is the simple way to run the bot (and stable), but the bot should always be started whenever you want to use /gpt. You can run yarn gpt-on-discord on your local PC or remote instance, to start the bot daemon that handles /gpt command.
Simple discord bot that uses OpenAI's GPT-3 API to generate natural language responses to user input.Ĭreated for my practice of Discord Bot / ChatGPT.
0 kommentar(er)
