Home Open Source Article

ByteDance Releases Stable-DiffCoder, Diffusion-Based Code LLM

TL;DR

ByteDance open-sources Stable-DiffCoder, an 8B diffusion-based code model that outperforms autoregressive baselines on code generation, completion, and editing tasks.

Key Points

  • Introduces block diffusion continual pretraining (CPT) with tailored warmup and block-wise clipped noise schedule
  • Achieves state-of-the-art performance among open-source 8B models across code generation, completion, editing, and reasoning benchmarks
  • Uses identical architecture and data as AR baseline for fair comparison, isolating diffusion training's impact
  • Publicly available on Hugging Face with three model variants; MIT licensed

Why It Matters

Demonstrates that diffusion-based training can improve code modeling quality beyond autoregressive approaches, even with controlled parameters. Provides actionable training insights and a production-ready alternative for developers building code generation systems at scale.
View on GitHub

Source: github.com