ZHZisZZ/dllm

[ICLR 2026] Official code for TraceRL: Revolutionizing post-training for Diffusion LLMs, powering the SOTA TraDo series.

The first continuous diffusion language model that rivals discrete counterparts on standard language modeling benchmarks like LM1B and OpenWebText.

Research code and optimized inference for ALoDLM: Adaptively Looped Diffusion Language Models.

[COLM 2026] Official repository of DARE: Diffusion Large Language Models Alignment and Reinforcement Executor

ORL: On-Policy Residual Learning for Diffusion Language Models

Frequently updated list of dLLM (Diffusion Large Language Models) papers, models, and other resources

Flexible and Pluggable Serving Engine for Diffusion LLMs

Easy and Efficient dLLM Fine-Tuning