This library implements constrained decoding (also called constrained sampling or structured outputs) for Large Langauge Models (LLMs). It can enforce arbitrary context-free grammar on the output of LLM and is fast - on the order of 1ms of CPU time per token (for 100k tokenizer) with negligible startup costs.
Following grammar formats are supported:
llguidance - internal (JSON-based) formatThe internal format is most powerful and can be generated by the following libraries:
The library can be used from:
The library is currently integrated in:
The integration is ongoing in:
Given a context-free grammar, a tokenizer, and a prefix of tokens, llguidance computes a token mask - a set of tokens from the tokenizer - that, when added to the current token prefix, can lead to a valid string in the language defined by the grammar. Mask computation takes approximately 1ms of single-core CPU time for a tokenizer with 100k tokens. While this timing depends on the exact grammar, it holds, for example, for grammars derived from JSON schemas. There is no significant startup cost.
The library implements a context-free grammar parser using Earley’s algorithm on top of a lexer based on derivatives of regular expressions. Mask computation is achieved by traversing the prefix tree (trie) of all possible tokens, leveraging highly optimized code.
LM-format-enforcer and llama.cpp grammars are similar to llguidance in that they dynamically build token masks for every step of the decoding process. Both are significantly slower - the former due to clean Python code and the latter due to the lack of a lexer and use of a backtracking parser, which, while elegant, is inefficient.
Outlines builds an automaton from constraints and then pre-computes token masks for all automaton states, making sampling fast but inherently limiting constraint complexity and introducing significant startup cost and memory overhead. Llguidance computes token masks on the fly and has essentially no startup cost. The lexer’s automata are built lazily and are typically much smaller, as the context-free grammar imposes the top-level structure.
In llguidance, online mask computation takes approximately 1ms of CPU time per sequence in a batch. Thus, with 16 cores and a 10ms forward pass, the library can handle batch sizes up to 160 without slowing down the model. (Note that a 10ms forward pass for small batch sizes typically increases to 20ms+ for batch sizes of 100-200.)
If you just need the C or Rust library (llguidance_parser),
check the parser directory.
For Python bindings:
./scripts/install-deps.sh./scripts/test-guidance.shThis builds the Python bindings for the library and runs the tests (which mostly live in the Guidance repo - it will clone it).
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.
When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.