Alternatives to Guidance

Open-source Python library that constrains LLM decoding so JSON, regex, and grammar outputs are valid on the first call.

Free, MIT-licensed OpenAI API examples and Python notebooks for developers building with the OpenAI API.

Open-source TypeScript AI SDK for streaming AI apps and agents across 100+ models with built-in fallbacks

Mirascope is an open-source Python library that turns LLM calls, tools, and prompt versioning into plain decorated functions.

Build production AI agents with open source Python and TypeScript harness SDKs you fully control.

Mellea turns type-annotated Python functions into validated, testable LLM calls with token-level constrained decoding.

A headless React hooks library that runs Vicuna-13B chat entirely in your browser via WebGPU, with no prompt or completion leaving the client.

Open-source Python library for loading, fine-tuning, and running transformer models across text, vision, audio, and video.

xsAI is an extra-small OpenAI-compatible AI SDK for JavaScript and TypeScript — 13.1 KB gzipped with every package included.

SwiftOpenAI is an MIT-licensed Swift package for calling the OpenAI API with native async/await and Combine on iOS, macOS, and visionOS.

Pydantic AI is the typed, open-source Python agent framework that returns Pydantic-validated objects from every run.

Open-source Java library that tokenizes text with OpenAI's BPE encodings so you can count, encode and decode locally

Open-source Rust library for building streaming, parallel LLM pipelines — data indexing, retrieval-augmented generation, and autonomous agents.

Build agentic AI apps in TypeScript, Go, Python, or Dart with Genkit's open-source AI SDK.

Go SDK for AI apps: one type-safe API for text generation, streaming, tools and MCP across 25+ LLM providers.

Free browser markdown editor with Battle Mode for comparing OpenAI, Claude, and Gemini prompt outputs, client-side only.

Open-source JavaScript library that runs small LLMs like Llama 3.2 1B Instruct inside the browser via WebAssembly and WebGPU.

Open Strawberry is a free, open-source chain-of-thought inspector for reading and comparing model reasoning traces.

Open-source Java library that puts a unified LLM and vector store API inside your JVM app

Predibase: managed fine-tuning and multi-LoRA serving for open-source LLMs, now inside Rubrik.

Open-source React framework for agent chat, generative UI, and shared state on any AG-UI backend.

Eino is an open-source Go LLM framework from ByteDance for building agents, graphs, and RAG pipelines in your existing Go backend.

Open-source LLM from UC Berkeley that turns natural language into API calls and tool-using agent actions.

OpenUI turns AI agent responses into streaming generative UI built from your own component library.

PrivateGPT is the open-source on-premise RAG framework for local document Q&A — 57k GitHub stars, zero cloud calls.

Rig is an open-source Rust LLM framework for building type-safe AI agents with one unified API across 20+ model providers.

Python agent framework for small language models — local engines, OpenAI-compatible servers, or hosted providers, with tool-call tracing

Open-source (MIT) MCP testing and full-stack MCP App framework for building ChatGPT Apps and Claude Connectors from one React codebase.

Open-source Ruby gem that calls 11 LLM providers through one consistent interface for embeddings, completions, and chat.

Open-source Ruby gem giving one unified API across Anthropic, DeepSeek, Google, Mistral, Llama, and OpenAI