Which AI providers does Appwrite integrate with?

Appwrite has tutorials and function templates for OpenAI, Hugging Face, Pinecone, ElevenLabs, Perplexity, Replicate, fal.ai, LangChain, Together AI, and Anyscale. New integrations are added regularly, and the full list lives in the [Appwrite AI documentation](/docs/products/ai).

What are AI Function templates?

AI Function templates are pre-built [Appwrite Functions](/docs/products/functions) that wire a third-party AI provider into your project. You pick a template in the Console (for example, **Prompt ChatGPT** or **Hugging Face inference**), deploy it, and the function exposes an endpoint your client can call. No infrastructure setup is required.

How do I add an OpenAI integration to my Appwrite app?

Use the OpenAI function template from the Console, supply your OpenAI API key as an environment variable, and deploy. Your client SDK can then call the function to run completions, chat, embeddings, or other OpenAI APIs without exposing the key to end users.

Can I use Appwrite for vector search?

Appwrite integrates with external vector databases like Pinecone and Upstash Vector through Function templates and tutorials. You typically store metadata in [Appwrite Databases](/docs/products/databases) and embeddings in the vector store, then call both from a server-side Function during queries.

What is a knowledge base in the context of AI apps?

In retrieval-augmented generation (RAG), a knowledge base is the collection of source documents and their embeddings that a model uses to ground its answers. With Appwrite, you can store source content in [Storage](/docs/products/storage) or Databases, generate embeddings inside a Function, and query a vector index at runtime.

Where can I see real-world AI apps built on Appwrite?

Visit [builtwith.appwrite.io](https://apwr.dev/yAtAG8j) to browse community-submitted projects, and check the [Appwrite blog](/blog) for tutorials such as personal chatbots, audio processing pipelines, and computer vision integrations using the available providers.

Announcing new Appwrite AI integrations

The AI hype is real and will be around for many years to come. In 2021 alone, AI startups worldwide raised nearly $50 billion in venture capital across approximately 1,500 deals, reflecting a significant increase from previous years. So it's no surprise that many of you are looking to build AI powered applications. But that's easier said than done, as building AI powered applications can be tricky. We don't want it to be. That's why we're happy to share multiple AI related announcements designed to enhance the Appwrite experience and adapt to new possibilities for devs building with Appwrite.

New knowledge base for AI
New third party AI integration tutorials
New AI powered Function templates

Let's examine the integrations, Function templates, and what's still to come.

AI use cases and documentation

To make your life easier, we have added technical guides to our documentation covering computer vision, natural language processing, and audio processing. These advanced tutorials explain how to use AI and Appwrite in your projects.

AI docs

AI integrations

We've also added support for many AI tools so you can integrate them into your projects. You will find step-by-step tutorials for each integration in our documentation. So far, we have added integrations and tutorials for:

AI powered Function templates

Function templates are pre-built Appwrite Function that can be integrated into your Appwrite project with just a few clicks. With templates, you can easily incorporate new features and integrations into your app without writing additional code or managing infrastructure.

We have templates for your favorite AI tools, like OpenAI, LangChain, HuggingFace, and many more, covering many use cases, such as building an intelligent chatbot using OpenAI.

Find the full overview on your console under the Templates tab in Fuctions. Filter function templates by the AI use case and select the AI function template that you need.

Appwrite console with AI template functions

You can also find the full list of templates on GitHub and see what's to come.

Building more AI features

Over the coming months, we will roll out more AI features to make building AI products smoother and allow more flexibility.

Vector database

Another advanced AI use case in the field of Natural Language Processing is the ability to use vector databases to store and process embeddings.

Embeddings are numerical representations of words, sentences etc., and vector databases are optimized to store and perform similarity searches on the embeddings.

Adding a vector database will be another important milestone for making Appwrite even more AI compatible.

AI marketplace

We will add an AI marketplace to encourage more open source collaboration on AI models and improve discoverability. The marketplace will allow you to build and share popular open-source and custom models. You can also host and sell your own custom models on the marketplace.

Building the future

The future of Appwrite is exciting, and we look forward to continuing to build exciting new features. Join us on Discord to participate in the discussion and have fun. For Appwrite and AI-powered projects, visit builtwith.appwrite.io.