BoundaryML/baml

Health v2 · maintenance only

Adoption not part of the score

9078 stars · 488 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

BAML is a domain-specific programming language for building LLM agents, with a Rust-like type system, typed errors, built-in tests/evals, and an agent stdlib. It compiles fast and can be called incrementally from TypeScript, Python, Go, C#, Java, and Rust.

Use cases

• define type-safe llm functions with structured output
• build ai agents with fewer mistakes
• add guardrails to llm calls
• write evals and tests for prompts
• call llm functions from python or typescript
• parse structured data from llm responses

When to choose

• you want strongly typed structured outputs from LLMs
• you are building agents and need built-in testing and evals
• you want a language-agnostic LLM layer callable from TS, Python, Go, Java, or Rust

When to avoid

• you prefer staying entirely in your host language without a DSL
• you need a general-purpose programming language rather than an LLM-focused one
• your project has no LLM or agent components

Facets

library · maturity active

programming-language llm-inference agent-framework prompt-engineering type-system testing large-language-models programming-languages developer-tools cross-platform python rust cli llm structured-output guardrails agents dsl type-safe evals ai-agents nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="BoundaryML/baml")