How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.
This is the official implementation of the paper d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillation (ICML'26), where we introduce a novel recipe for building an ultra-fast diffusion language model named d3LLM (pseuDo-Distilled Diffusion LLM) 🚀.
The main features of hao-ai-lab/d3llm are: Training and Alignment.
Projects with overlapping indexed features include: amap-ml/ar-map — Are Autoregressive Large Language Models Implicit Teachers for Diffusion Large Language Models? A comprehensive… apple/ml-diffucoder — This software project accompanies the research paper, DiffuCoder: Understanding and Improving Masked Diffusion Models… autonomousvision/mdpo — [[Paper]](https://arxiv.org/pdf/2508.13148) [[Project]](https://cli212.github.io/MDPO/). bhmingyang/tad — ```bash conda create -n tad python=3.10 -y conda activate tad. bowen-sun-0728/blockwise-sft — Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding. aistairc/text-diff-2step-loss — The implementation of "Addressing the Training-Inference Discrepancy in Discrete Diffusion for Text Generation".
Are Autoregressive Large Language Models Implicit Teachers for Diffusion Large Language Models? A comprehensive framework for transferring alignment knowledge from AR-LLMs to Diffusion Models
This software project accompanies the research paper, DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation.
[Paper](https://arxiv.org/pdf/2508.13148) [Project](https://cli212.github.io/MDPO/)
The implementation of "Addressing the Training-Inference Discrepancy in Discrete Diffusion for Text Generation"