eth-sri/constrained-diffusion
Constrained Decoding of Diffusion LLMs with Context-Free Grammars.
This tool helps developers generate text, code, or chemical structures using diffusion Large Language Models (LLMs) like LLaDA or DiffuCoder. It takes a prompt and a formal grammar (like JSON Schema, C++, or SMILES) and produces output that strictly adheres to the specified grammar. This is useful for developers who need to ensure the generated content is syntactically correct and functionally robust.
Use this if you are a developer generating code, structured data (like JSON), or chemical notations and need to guarantee the output from a diffusion LLM follows specific syntactic rules.
Not ideal if you are not a developer and do not work directly with Large Language Models or formal grammars.
Stars
41
Forks
5
Language
Rust
License
MIT
Category
Last pushed
Dec 17, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/eth-sri/constrained-diffusion"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ljleb/sd-mecha
Executable State Dict Recipes
SJTU-DENG-Lab/Discrete-Diffusion-Forcing
Discrete Diffusion Forcing (D2F): dLLMs Can Do Faster-Than-AR Inference
declare-lab/tango
A family of diffusion models for text-to-audio generation.
Li-Jinsong/DAEDAL
[ICLR 2026] Official repository of "Beyond Fixed: Training-Free Variable-Length Denoising for...
SalesforceAIResearch/CoDA
Salesforce AI Research's open diffusion language model