Alternatives to Marvin

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

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

Native PHP SDK for LLM inference, typed object extraction, agents, and coding-agent control.

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

BAML is a type-safe language with an agent-ready CLI for building, testing, tracing and querying LLM functions.

Guidance is a Python constrained generation library that steers LLMs with regex, context-free grammars, and inline control flow.

One strongly-typed C# .NET client for calling 15+ LLM providers through a unified TornadoApi.

Genkit is Google's open-source AI SDK for building agentic AI apps in TypeScript, Go, Python, and Dart.

Fabric turns reusable AI prompt patterns into chainable command-line workflows you can run, inspect, and version in your terminal.

Local MCP context gateway that compresses and filters what your AI coding agent reads

TypeLLM turns text and images into typed, schema-conformant values in one workflow API call — no JSON repair loops.

CopilotKit is an open-source React framework for adding agent chat, generative UI, and human approvals to your app.

Outlines enforces structured generation by constraining LLM decoding so JSON, regex, and grammar output is valid at every token.

Tiptap AI Toolkit lets your AI agent edit rich-text documents in place with schema-aware, reviewable patches.

Open-source React toolkit for generative UI that renders your own components from natural language

Open-source TypeScript AI SDK for streaming AI apps and agents with multi-model support and built-in fallbacks

OpenUI is the open standard for generative UI: an LLM writes OpenUI Lang and your registered components render as live interfaces.

Headroom builds custom AI software that reads your email, documents and calls and turns them into records your team runs on.

Promptmetheus is a prompt engineering IDE for composing, testing, and scoring prompts across 150+ LLMs from 15 providers.

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

Open Strawberry lets you inspect and compare chain-of-thought reasoning traces across six LLM backends, free.

flompt is a free visual prompt builder that compiles prompts into Claude-optimized XML blocks

OpenAI Cookbook: free, MIT-licensed notebooks and guides for calling the OpenAI API

Open-source React chat UI components for building LLM chat interfaces in hours, not weeks

Open-source Java library giving JVM apps one unified API for LLMs and vector stores

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

LMQL is a programming language for LLMs that uses types, templates and constraints to make prompts modular and outputs reliable.

OpenAI Scala Client: typed, effect-friendly JVM calls to OpenAI's chat, image, embedding and audio APIs

Open-source JS/TS toolkit for building multimodal LLM agents with streaming, validation, and managed memory

DiscuroAI chains GPT-3 prompt templates and DALL-E 2 image calls into multi-step workflows you run with one API request that returns JSON.