GerbilGerbil
DocsHarnessPlaygroundModelsIntegrations
npmGet Started
Documentation
  • How It Works
  • React Hooks
  • Semantic Search
  • Context Tester
  • Getting Started
  • Browser
  • Mobile & PWA
  • Models
  • Vision
  • Text-to-Speech
  • Speech-to-Text
  • Caching
  • Persistence
  • Concurrency
  • Observability
  • Embeddings
  • Memory & RAG
  • Skills
  • Tools
  • Structured Output
  • Autocomplete
  • Frameworks
    • Vercel AI SDK
    • Next.js
    • React
    • Express
    • Hono
    • LangChain
    • LlamaIndex
  • MCP Server
  • CLI
  • REPL
  • API Reference

Framework Integrations

Gerbil drops into the frameworks you already use, from the Vercel AI SDK to Next.js and beyond. Wherever JavaScript runs, you can run local inference alongside it.

Vercel AI SDK

Vercel AI SDK

Drop-in provider for generateText, streamText, and structured output. The recommended way to use Gerbil in production apps.

StreamingTool callingStructured outputThinking mode
→
Next.js

Next.js

API route handlers, React Server Components, and client hooks for Next.js 13+ App Router.

API routesServer ComponentsClient hooksStreaming
→
React

React

Client-side hooks for generation, streaming, and chat interfaces. Works with any React framework.

useGerbiluseChatStreaming UIError handling
→
Express

Express

Middleware and route handlers for Express.js applications. Mount all AI routes with a single line.

MiddlewareREST APISSE streamingEmbeddings
→
Hono

Hono

Edge-ready middleware for Hono. Works with Cloudflare Workers, Deno, Bun, and Node.js.

Edge runtimeBun supportDeno supportStreaming
→
LangChain

LangChain

LLM and embeddings for LangChain. Build chains, agents, and RAG pipelines with local models.

ChainsAgentsVector storesRAG
→
LlamaIndex

LlamaIndex

LLM and embedding integration for LlamaIndex. Build RAG applications with local models.

Query enginesChat enginesDocument loadersPersistence
→

Don't see your framework?

Gerbil works with any JavaScript framework. Check out the Getting Started guide to use the core API directly.

Want to contribute an integration? Open a PR on GitHub →

On this page
GerbilGerbil

Local LLM inference

Product

  • Getting Started
  • Playground
  • Skills
  • Models

Integrations

  • AI SDK
  • Next.js
  • Express
  • MCP Server

Resources

  • API Reference
  • GitHub
  • npm
  • Discord

Ecosystem

  • Taskmaster
  • Hamster

2026 Wheel Go Fast / MIT License

// Built by Hamster with Hamster