# genkit-ai/genkit

Open-source framework for building agentic apps in JavaScript, Go, Dart, and Python, built and used in production by Google

Repository: https://github.com/genkit-ai/genkit
Canonical: https://ross.abutalabs.com/products/genkit
Homepage: https://genkit.dev
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: ai, genkit, vector-database, embedders, agents, llm, multimodal, rag
Last push: 2026-08-26T23:09:29+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 61
- inputs: {"age_days": 856, "days_push": 7, "days_rel": 8, "gap_med": 4, "n_releases_24m": 126}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6378, forks 837 (observed 2026-08-28T04:09:42.930775+00:00)

## What it is
Genkit is an open-source framework by Google for building full-stack, AI-powered agentic applications in JavaScript/TypeScript, Go, Python, and Dart. It provides a unified, model-agnostic API for generation, structured output, tool calling, RAG, and multimodal content, plus a CLI and Developer UI for testing, tracing, and deployment.

## Use cases
- build a chatbot with llm tool calling
- add rag with a vector database to my app
- generate structured  output from an llm
- swap between openai, anthropic, and gemini models with one api
- build an ai agent workflow in go or python
- deploy ai flows as http endpoints
- debug and trace llm calls locally with a developer ui

## When to choose
- you want a production-backed, multi-language framework for AI features
- you need model-agnostic generation with easy provider swapping
- you're building full-stack agentic apps with Next.js, React, Angular, Flutter, or standalone backends
- you want built-in observability, tracing, and local dev tooling for AI workflows

## When to avoid
- you only need a thin single-provider SDK without workflow abstractions
- you need mature Python or Dart support today (still beta/preview)
- you want a framework-locked-to-one-cloud experience or non-AI application logic only

## Facets
- artifact type: framework
- maturity: stable
- function: agent-framework, llm-inference, rag, prompt-engineering, chatbot, sdk, plugin-system, web-framework
- domain: artificial-intelligence, large-language-models, chatbots, web-development, developer-tools
- platform: go, python, cross-platform, cli
- tags: generative-ai, multi-language, tool-calling, structured-output, multimodal, model-agnostic, observability, flows, google, firebase, ai-agents, retrieval-augmented-generation, nodejs, web-server

## Member repositories
- genkit-ai/genkit (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:42.930775+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:45:10.048409+00:00, confidence not recorded.
  - readme: https://github.com/genkit-ai/genkit (fetched 2026-08-28T04:09:42.930775+00:00, sha aa3636eb8e13)
  - homepage: https://genkit.dev (fetched 2026-08-29T08:42:31.735280+00:00, sha ee3fef2e69e3)
  - site_page: https://genkit.dev/docs/get-started (fetched 2026-08-29T08:42:31.744514+00:00, sha be2727683b07)
  - site_page: https://genkit.dev/docs/roadmap (fetched 2026-08-29T08:42:31.746463+00:00, sha 1d49c128ac99)
  - site_page: https://genkit.dev/docs/js/get-started (fetched 2026-08-29T08:42:31.748313+00:00, sha be2727683b07)
  - site_page: https://genkit.dev/docs/models (fetched 2026-08-29T08:42:31.749968+00:00, sha 013c49516b90)
  - site_page: https://genkit.dev/docs/flows (fetched 2026-08-29T08:42:31.752442+00:00, sha c7e4357f3cb8)
  - site_page: https://genkit.dev/docs/devtools (fetched 2026-08-29T08:42:31.754214+00:00, sha c734ea2fe05c)
- Data as of 2026-08-30T08:39:29.467469+00:00.