llguidance is a constrained-decoding engine written in Rust, maintained under the guidance-ai organization. It takes a grammar or JSON Schema and computes, at each decoding step, the set of tokens that keep the output valid, applying that as a mask on the sampler. It is built to be the low-level enforcement layer that a higher-level library or a serving engine calls, rather than a tool a developer uses directly. It is the engine underneath Guidance, and is one of the interchangeable constrained-decoding backends alongside Outlines and XGrammar. The split between these libraries and the developer-facing tools above them (Guidance, Instructor) is the main thing to understand: the same masking idea has both an ergonomic API surface and a performance-critical core, and llguidance is squarely the core. For open-source AI it reflects the maturing of structured output from a Python-level trick into a dedicated, low-overhead systems component, which is what lets open engines offer schema enforcement without giving up throughput.
The Stack · Runtime · Open source
llguidance
Low-level Rust constrained-decoding engine that enforces grammars and JSON Schema at high throughput; powers Guidance.
Sources
- llguidance on GitHub https://github.com/guidance-ai/llguidance
Want a follow-up? Ask the chat about llguidance in context. It will compare to siblings at the same layer and ground every claim in the wiki.
Other projects at the Runtime layer
12 siblings · ordered open first
- vLLM Open source
Dominant open production inference engine; PagedAttention and continuous batching; NVIDIA / AMD / Intel / TPU support.
- SGLang Open source
RadixAttention plus structured generation; from the LMSYS team; gains for shared-prefix and agent workloads.
- llama.cpp Open source
Georgi Gerganov's local-first inference engine; defines the GGUF format; the on-device standard.
- Ollama Open source
Local model runner; Docker-style UX over llama.cpp; the easiest way to run open weights on your machine.
- Text Generation Inference (TGI) Open source
HuggingFace's production inference server; maintenance mode in 2026 as vLLM became the standard.
- MLC-LLM Open source
Cross-platform compilation (TVM-based); the 'LLM in your browser' or 'on your phone' standard.
- Outlines Open source
Structured-output library using finite-state-machine guided decoding to force model output to match a regex, JSON Schema, or grammar.
- XGrammar Open source
Fast grammar-based constrained-decoding engine, used as a structured-output backend by several open serving engines.
- Guidance Open source
A programming model for constrained generation that interleaves control flow with model calls and enforces regex, grammars, and JSON Schema.
- Instructor Open source
Library for getting Pydantic-typed structured output from LLMs across many providers, built on function calling and JSON mode with automatic retries.
- LM Format Enforcer Open source
Token-filtering library that enforces a JSON Schema or regex during generation; integrates with Transformers, vLLM, and llama.cpp.
- TensorRT-LLM Source available
NVIDIA's closed-runtime counterpart; fastest on NVIDIA hardware; depends on closed CUDA kernels and the proprietary TensorRT compiler.