Guidance
Microsoft Research library for constrained LLM generation with a Python guidance language, token healing and multi-backend support.
Tags
Description
Established open-source structured-generation toolkit (not a 2026 launch). Install guidance from PyPI (0.3.1 verified 2026-11-09). Interleaves generation, prompting and control flow; supports constrained decoding patterns across supported model backends. MIT license (guidance-ai/guidance; GitHub not archived). Model-provider credentials and compute are separate from the free library. Companion llguidance offers a faster structured-output engine used by some stacks.
Agent access
Related
- Outlines: Structured LLM output library from .txt (dottxt): JSON/schema/grammar constraints with broad model integrations.
- llguidance: High-performance structured-output / constrained-decoding engine from the Guidance project (Rust core, Python bindings).
- XGrammar: Fast portable structured-generation engine for LLM grammars and constrained decoding (MLC AI).
Guides & comparisons
- Ollama vs llama.cpp for agents — tools, OpenAI APIs, grammars and local serving (Comparison): Compare Ollama and llama.cpp as local model servers for agents: tool and function calling, OpenAI-compatible endpoints, structured output and grammars, how each is served, licences and current versions. Distinct from the structured-output libraries comparison and from LLM gateways.
- Structured-output libraries compared — Outlines, XGrammar, Guidance and llguidance (Comparison): Compare Outlines, XGrammar, Guidance and llguidance by constraint form (regex, JSON Schema, CFG), documented engine integrations, licences, current PyPI versions and official JSON Schema coverage limits. The libraries compose: Outlines and Guidance can sit on these backends.
Sources
- https://github.com/guidance-ai/guidance
- https://github.com/guidance-ai/guidance/blob/main/LICENSE.md
- https://pypi.org/project/guidance/
- https://guidance.readthedocs.io/
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)