Skip to content
forked from ykhli/AI-tamago

A local-ready LLM-generated and LLM-driven virtual pet with thoughts and feelings. 100% Javascript.

License

Notifications You must be signed in to change notification settings

twelvemo/AI-tamago

 
 

Repository files navigation

AI Tamago 🥚🐣

A 100% local, LLM-generated and driven virtual pet with thoughts, feelings and feedback. Revive your fond memories of Tamagotchi! https://ai-tamago.fly.dev/

All ascii animations are generated using chatgpt (included prompts in the repo).

Have questions? Join AI Stack devs and find me in #ai-tamago channel.

Demo 🪄

ai-tamago-demo.mov

Stack

Local Mode

Prod Mode

All of above, plus:

  • 🔐 Auth & User Management: Clerk
  • ☁️ Hosting: Fly
  • 🥇 Rate Limiting: Upstash

Overview

Prerequisites

Quickstart

1. Fork and Clone repo

Fork the repo to your Github account, then run the following command to clone the repo:

git clone git@github.com:[YOUR_GITHUB_ACCOUNT_NAME]/AI-tamago.git

2. Install dependencies

cd ai-tamago
npm install

All client side tamagotchi code is in Tamagotchi.tsx

3. Install Ollama

Instructions are here.

4. Run Supabase locally

  1. Install Supabase CLI
brew install supabase/tap/supabase
  1. Start Supabase

Make sure you are under /ai-tamago directory and run:

supabase start

Tips: To run migrations or reset database -- seed.sql and migrations will run supabase db reset

5. Fill in secrets

Note: The secrets here are for your local supabase instance

cp .env.local.example .env.local

Then get SUPABASE_PRIVATE_KEY by running

supabase status

Copy service_role key and save it as SUPABASE_PRIVATE_KEY in .env.local

6. Set up Inngest

npx inngest-cli@latest dev

Make sure your app is up and running -- Inngest functions (which are used to drive game state) should register automatically.

7. Run app locally

Now you are ready to test out the app locally! To do this, simply run npm run dev under the project root and visit http://localhost:3000.

Deployment Guide

Now you have played with the AI tamago locally -- it's time to deploy it somewhere more permanent so you can access it anytime!

0. Choose which model you want to use in production

  • If you want to test out using Chatgpt in prod, simply remove LLM_MODEL=ollama from .env.local and fill in OPENAI_API_KEY
  • If you want to try Replicate, set LLM_MODEL=replicate_llama and fill in REPLICATE_API_TOKEN
  • If you want to deploy Ollama yourself, you can follow this awesome guide -- Scaling Large Language Models to zero with Ollama. It is possible to run Ollama on a performance-4x Fly VM (CPU) with 100gb volume, but if you can get access to GPUs they are much faster. Join Fly's GPU waitlist here if you don't yet have access!

1. Switch to deploy branch -- this branch includes everything you need to deploy an app like this.

git co deploy

This branch contains a multi-tenancy-ready (thanks to Clerk) app, which means every user can get their own AI-tamago, and has token limit built in -- you can set how many times a user can send requests in the app (see ratelimit.ts)

2. Move to Supabase Cloud:

  • Create a Supabase project here, then go to Project Settings -> API. Fill out secrets in .env.local
  • SUPABASE_URL is the URL value under "Project URL"
  • SUPABASE_PRIVATE_KEY is the key starts with ey under Project API Keys
  • Copy Supabase project id, which you can find from the url https://supabase.com/dashboard/project/[project-id]

From your Ai-tamago project root, run:

supabase link --project-ref [insert project-id]
supabase migration up
supabase db reset --linked

3. Create Upstash Redis instance for rate limiting

This will make sure no one user calls any API too many times and taking up all the inference workloads. We are using Upstash's awesome rate limiting SDK here.

  • Sign in to Upstash
  • Under "Redis" on the top nav, click on "Create Database"
  • Give it a name, and then select regions and other options based on your preference. Click on "Create"
  • Scroll down to "REST API" section and click on ".env". Now you can copy paste both environment variables (UPSTASH_REDIS_REST_URL and UPSTASH_REDIS_REST_TOKEN) to your .env.local

4. Now you are ready to deploy everything on Fly.io!

  • Register an account on fly.io and then install flyctl
  • Run fly launch under project root. This will generate a fly.toml that includes all the configurations you will need
  • Run fly scale memory 512 to scale up the fly vm memory for this app.
  • Run fly deploy --ha=false to deploy the app. The --ha flag makes sure fly only spins up one instance, which is included in the free plan.
  • For any other non-localhost environment, the existing Clerk development instance should continue to work. You can upload the secrets to Fly by running cat .env.local | fly secrets import
  • If you want to make this a real product, you should create a prod environment under the current Clerk instance. For more details on deploying a production app with Clerk, check out their documentation here. Note that you will likely need to manage your own domain and do domain verification as part of the process.
  • Create a new file .env.prod locally and fill in all the production-environment secrets. Remember to update NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY and CLERK_SECRET_KEY by copying secrets from Clerk's production instance -cat .env.prod | fly secrets import to upload secrets.

If you have questions, join AI Stack devs and find me in #ai-tamago channel.

Other Resources

About

A local-ready LLM-generated and LLM-driven virtual pet with thoughts and feelings. 100% Javascript.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • TypeScript 85.0%
  • CSS 8.5%
  • PLpgSQL 2.5%
  • JavaScript 2.4%
  • Dockerfile 1.6%