Repolode
Submit
genkit-ai

genkit-ai/genkit

Open-source framework for building full-stack AI apps in JavaScript/TypeScript, Go, Python, and Dart, with unified model provider APIs.

View on GitHubWebsite
Stars
6,572
Forks
878
Open issues
861
Last commit
24 hours ago

What is it?

Genkit is a framework for building AI-powered applications, originally built and used in production by Google's Firebase team. It offers a single interface for calling models from providers such as Google, OpenAI, Anthropic, and Ollama, so application code does not need to change when you switch providers.

The framework covers common building blocks for LLM apps: text and image generation, type-safe structured output, tool calling, prompt templates, persisted chat, multi-step workflows (flows), and retrieval-augmented generation (RAG). The JavaScript/TypeScript and Go SDKs are marked production-ready, while Python is in beta and Dart is in preview.

A local CLI and Developer UI let you run prompts and flows, inspect execution traces, and run evaluations against datasets. Apps can be deployed to Cloud Functions for Firebase, Google Cloud Run, or other environments that support the chosen language, and client SDKs are provided for Next.js, React, Angular, iOS, and Android.

Key features

  • 01Unified API for models from Google, OpenAI, Anthropic, Ollama, and others
  • 02Type-safe structured output and tool calling
  • 03Prompt templating, persisted chat, and multi-step flows
  • 04Retrieval-augmented generation (RAG) support
  • 05Local CLI and Developer UI for testing, tracing, and evaluation

Use cases

  • Building chatbots or assistants that need to switch between model providers
  • Adding RAG-based question answering over your own data to a web app
  • Creating agentic workflows with tool calls in TypeScript or Go backends
  • Shipping AI features to Firebase, Cloud Run, or mobile apps with client SDKs

Quick start

  1. 1

    Install the Genkit CLI

    npm install -g genkit-cli
  2. 2

    Get an API key from a model provider such as Google Gemini

  3. 3

    Initialize Genkit with your provider plugin in your code, as shown in the README example

  4. 4

    Run your code wrapped with the Genkit CLI to open the developer UI

    genkit start -- <command to run your code>

Things to know

  • The Python SDK is in Beta and the Dart SDK is in Preview.
  • Using model providers requires an API key from that provider.

In collections

Open-source AI agent frameworks & apps

Topics

#llm#ai-framework#rag#agents#typescript#go#tool-calling#vector-database#ai#embedders

FAQ

What is genkit?
Open-source framework for building full-stack AI apps in JavaScript/TypeScript, Go, Python, and Dart, with unified model provider APIs. Genkit is a framework for building AI-powered applications, originally built and used in production by Google's Firebase team. It offers a single interface for calling models from providers such as Google, OpenAI, Anthropic, and Ollama, so application code does not need to change when you switch providers.
Is genkit free to use? What license does it have?
genkit is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is genkit still maintained?
Actively maintained. The last commit to genkit was on Nov 6, 2026.
What can I use genkit for?
Building chatbots or assistants that need to switch between model providers Adding RAG-based question answering over your own data to a web app Creating agentic workflows with tool calls in TypeScript or Go backends Shipping AI features to Firebase, Cloud Run, or mobile apps with client SDKs
How do I install genkit?
Install the Genkit CLI: npm install -g genkit-cli Get an API key from a model provider such as Google Gemini Initialize Genkit with your provider plugin in your code, as shown in the README example Run your code wrapped with the Genkit CLI to open the developer UI: genkit start -- <command to run your code>
Where can I get genkit?
The source code and installation instructions are on GitHub: https://github.com/genkit-ai/genkit

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects