Stable-DiffCoder is a new code diffusion large language model that improves coding tasks using a unique training approach. It outperforms traditional autoregressive models on various benchmarks and is available for use on Hugging Face.

The article presents the Decoupled Diffusion Transformer (DDT) architecture, demonstrating improved performance with a larger encoder in a diffusion model framework. It achieves state-of-the-art FID scores on ImageNet benchmarks and allows for accelerated inference by reusing encoders across steps. The implementation provides detailed configurations for training and inference, along with online demos.