Llama 3 with Vercel AI

Use Llama 3 with Ollama to stream chat completions from Supabase Edge Functions using the Vercel AI SDK in Next.js.
* Blogpost: supabase.link/ai-blog-yt
* Docs: supabase.link/ef-ai-llm
* Find the code: git.new/vercel-ai-llama3
* Request access: supabase.link/ai-llm-early-ac...
Presented by Thor Schaeff (@thorwebdev go.thor.bio/x)
CHAPTERS:
00:00 Intro
00:32 Large Language Models
00:50 Vercel AI Chatbot demo
01:30 Large Language Models in Edge Functions
01:40 Install Ollama
02:16 Run Llama 3 locally with Ollama
03:30 Ollama serve
04:30 Initialise Supabase in Next.js app
05:00 Create a new Edge Function
06:00 Run llama3 prompt from Edge Function
07:47 Start Supabase locally
07:59 Run Edge Functions locally
08:20 Test Llama 3 model locally
08:47 Supabase Rap
09:48 Vercel AI useChat React Hook
10:27 OpenAI Chat Completion API route
12:59 Look at database in local Supabase Studio
13:51 Replicate OpenAI chat completion with Llama 3
18:25 Run the Vercel AI Chatbot demo locally
18:42 Vercel AI Chatbot with Llama3 demo
21:52 Request early access
💻 Videos to watch next:
▶ Why you should use Edge functions for AI: • Why you should use Edg...
▶ The missing pieces to your AI app (pgvector + RAG in prod): • The missing pieces to ...
▶ Build AI Apps FREE (& Start Earning Money): • Build AI Apps FREE - ...
👇 Learn more about Supabase 👇
🕸 Website: supabase.com/
🏁 Get started: app.supabase.com/
📄 Docs: supabase.com/docs
🔔 Subscribe for more tutorials and feature updates from Supabase: / @supabase
📱 Connect with Us:
🐙 Github: www.github.com/supabase
💬 Discord: www.discord.supabase.com/
🐦 Twitter: / supabase
▶ Instagram (follow for memes): / supabasecom
ABOUT SUPABASE:
Supabase is the open source Firebase alternative. Supabase provides a full Postgres database for every project with pgvector, backups, realtime, and more. Add and manage email and password, passwordless, OAuth, and mobile logins to your project through a suite of identity providers and APIs.
Build in a weekend, scale to millions.
#Supabase #AppDevelopment #RealtimeApps #DeveloperTools #AI #Ollama #Llama3

[@Supabase]

Thanks for tuning in \o/ Read the Launch Week Blogpost here: supabase.link/ai-blog-yt and then dive into the docs: supabase.link/ef-ai-llm 🚀

[@developerpranav]

first imma learn about semantic search!

[@hey_james]

Amazing stuff. Any chance you can stop ending every sentence in a higher pitch? 😂

[@mikejohneviota9293]

😂😂😂

[@anggoran7398]

idk but it happens unconsciously for several js devs...

[@thorwebdev]

Sorry, not an AI model, your prompt didn't work, haha 😅

[@didiercatz]

So, say we deploy this to prod, wouldn't the models run very slowly? Given that Supabase doesn't have beefy GPUs or w/e

[@twitchizle]

thats not how it works

[@didiercatz]

@@twitchizle so how does it work? Genuinely asking

[@Supabase]

You can either deploy your own production Ollama server, or you can submit the form to join our private alpha. Details in the docs: supabase.link/ef-ai-llm

[@champechilufya1458]

Whats the point of this ? What are you achieving or gaining from doing all this? Genuinely asking