Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference
By: Yuxuan Song , Zheng Zhang , Cheng Luo and more
Potential Business Impact:
Generates code super fast with top quality.
We present Seed Diffusion Preview, a large-scale language model based on discrete-state diffusion, offering remarkably fast inference speed. Thanks to non-sequential, parallel generation, discrete diffusion models provide a notable speedup to mitigate the inherent latency of token-by-token decoding, as demonstrated recently (e.g., Mercury Coder, Gemini Diffusion). Seed Diffusion Preview achieves an inference speed of 2,146 token/s over H20 GPUs while maintaining competitive performance across a sweep of standard code evaluation benchmarks, significantly faster than contemporary Mercury and Gemini Diffusion, establishing new state of the art on the speed-quality Pareto frontier for code models.
Similar Papers
Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
Machine Learning (CS)
Makes AI write and understand faster.
Planned Diffusion
Artificial Intelligence
Makes AI write faster without losing quality.
The Best of Both Worlds: Integrating Language Models and Diffusion Models for Video Generation
CV and Pattern Recognition
Makes videos from words, better than others.