Google's latest DiffusionGemma open AI model comes with a 4x speed boost
Another day, another AI model from Google. This time, Google DeepMind has released a new member of the Gemma 4 open model family , but it's fundamentally different from the rest of the lineup. DiffusionGemma doesn't generate outputs linearly like most AI models. Instead, it can produce an entire block of text in parallel. Google says this makes it faster and more efficient when running on local hardware like an Nvidia DGX or a humble gaming GPU. Most AI models are designed to
Google DeepMind has introduced DiffusionGemma, a new open AI model that diverges from traditional text generation methods. Unlike autoregressive models that produce text sequentially, DiffusionGemma generates entire blocks of text in parallel, similar to image diffusion models. This parallel processing approach significantly enhances its speed and efficiency, particularly on local hardware such as gaming GPUs and AI accelerators.
The model boasts a 26 billion parameter Mixture of Experts architecture, with a portion activated during inference, allowing it to fit within common GPU memory constraints. In benchmarks, DiffusionGemma demonstrated a speed increase of approximately four times compared to similarly sized autoregressive Gemma models, achieving over 1,000 tokens per second on high-end hardware.
This development offers a faster and more efficient method for text generation on consumer-grade hardware, potentially democratizing advanced AI capabilities.
📌 Kaynak
Bu özet Ars Technica kaynağından otomatik derlenmiştir. Tamamı için orijinal habere gidin.
Orijinal haberi oku →