# vercel/modelfusion

The TypeScript library for building AI applications.

Repository: https://github.com/vercel/modelfusion
Canonical: https://ross.abutalabs.com/products/modelfusion
Homepage: https://modelfusion.dev
Language: TypeScript
License: MIT
License Family: permissive
Topics: chatbot, gpt-3, javascript, js, llm, openai, ts, typescript, whisper, ai, embedding, huggingface, dall-e, stable-diffusion, llamacpp, artificial-intelligence, claude, multi-modal, ollama, mistral
Archived: true
Last push: 2024-07-19T15:17:19+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 85
- inputs: {"age_days": 1196, "days_push": 775, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1319, forks 95 (observed 2026-08-28T04:04:21.358598+00:00)

## What it is
ModelFusion is a TypeScript library that provides a unified, vendor-neutral abstraction layer for integrating AI models into JavaScript and TypeScript applications. It supports text streaming, object generation, tool usage, embeddings, and multi-modal models with production features like observability, retries, and type validation.

## Use cases
- build a chatbot in typescript
- stream llm text responses in a node app
- generate structured objects from llm output
- add tool calling to an ai agent
- generate embeddings for semantic search
- transcribe audio with whisper in javascript
- switch between openai, ollama, and llama.cpp providers
- build ai apps with retries and logging

## When to choose
- you want a typed, vendor-neutral abstraction over multiple AI model providers in TypeScript
- you need production features like retries, throttling, and observability hooks
- you want multi-modal support including text, image, speech, and embeddings in one library

## When to avoid
- you are starting a new project - the project is being integrated into the Vercel AI SDK, which is the recommended path forward
- you need a Python AI library
- you want a full application framework rather than a model integration layer

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, agent-framework, rag, chatbot, sdk, http-client, speech-recognition, tts, image-processing, prompt-engineering
- domain: artificial-intelligence, large-language-models, chatbots, web-development, developer-tools
- platform: cross-platform, serverless
- tags: multi-modal, openai, huggingface, ollama, llamacpp, text-streaming, tool-calls, embeddings, vercel-ai-sdk, vendor-neutral, ai-agents, natural-language-processing, nodejs, typescript, javascript

## Member repositories
- vercel/modelfusion (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:21.358598+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 2026-08-30T04:47:55.900463+00:00, confidence not recorded.
  - readme: https://github.com/vercel/modelfusion (fetched 2026-08-28T04:04:21.358598+00:00, sha c33db8dec955)
- Data as of 2026-08-30T08:39:29.467469+00:00.