DiffusionGemma: 4x faster text generation
Today, we’re introducing DiffusionGemma, an experimental open model that explores text diffusion, an exceptionally fast approach to text generation.
Key points
- Released under an Apache 2.0 license, this 26B Mixture of Experts (MoE) model moves beyond the sequential token-by-token processing of typical autoregressive Large Language Models (LLMs).
- While autoregressive Gemma 4 models remain the standard for high-quality production outputs, DiffusionGemma is designed for researchers and developers exploring speed-critical, interactive local workflows such as in-line editing, rapid iteration, and generating non-linear text structures.
- Blazing fast inference: By shifting the decode bottleneck from memory-bandwidth to compute, DiffusionGemma generates up to 4x faster token output on dedicated GPUs.
- Accessible hardware footprint: Operating as a 26B total Mixture of Experts (MoE) model that activates only 3.8B parameters during inference, DiffusionGemma fits comfortably within 18GB VRAM limits of high-end dedicated consumer GPUs when quantized.
Sources (1)
- [1]DiffusionGemma: 4x faster text generationGoogle DeepMind Blog · Jun 10, 04:24 PM
Today, we’re introducing DiffusionGemma, an experimental open model that explores text diffusion, an exceptionally fast approach to text generation.
Released under an Apache 2.0 license, this 26B Mixture of Experts (MoE) model moves beyond the sequential token-by-token processing of typical autoregressive Large Language Models (LLMs).
Extractive summary: sentences quoted from the sources.