文本扩散模型把生成速度拉到新高度，做代码补全或实时编辑的开发者可以直接在 NVIDIA 端点试跑，感受并行 token 的爽感。
Google DeepMind 推出实验性开源模型 DiffusionGemma，采用文本扩散技术，每步并行生成 256 个 token，推理速度可达 150+ TPS（DGX Spark）或 1000+ TPS（单张 H100）。该模型激活仅 3.8B 参数，量化后可在 24GB VRAM 消费级 GPU 上运行，适合代码填充、内联编辑等非线性任务。NVIDIA 从首日起提供 BF16/NVFP4 检查点、免费 GPU 加速端点及 vLLM 支持。DiffusionGemma 优先速度而非极致质量，生产场景仍推荐标准 Gemma 4。
Congrats to @GoogleDeepMind on the launch of DiffusionGemma. The model generates 256 tokens in para...
Congrats to @GoogleDeepMind on the launch of DiffusionGemma. The model generates 256 tokens in parallel per step, delivering 150+ TPS on DGX Spark, and 1,000+ TPS on a single H100. We're supporting it from day one with: • BF16 and NVFP4 checkpoints on @huggingface 🤗 • Free GPU-accelerated endpoints on build.nvidia.com S • @vllm_project support with FP8 precision Get started with DiffusionGemma on NVIDIA: nvda.ws/43ro19u s Google AI Developers @googleaidevs DiffusionGemma, our experimental open model released under an Apache 2.0 license, explores text diffusion, an exceptionally fast approach to text generation. Here’s how DiffusionGemma accelerates development: + Faster token output: By shifting the bottleneck from memory bandwidth to raw compute, the model generates up to 4x faster token output on dedicated GPUs + Accessible hardware footprint: Activates just 3.8B parameters during inference, fitting comfortably within 24GB-VRAM high-end consumer GPUs when quantized + Novel workflows: Parallel token generation enables self-correction, making it ideal for code infilling, in-line editing, and non-linear structures DiffusionGemma prioritizes speed over raw quality and accelerates best on compute-bound hardware (like @NVIDIAAI GPUs). Standard @GoogleGemma 4 remains recommended for production quality and memory-bound devices. 🔗 View Quoted Tweet 💬 8 🔄 4 ❤️ 20 👀 1174 📊 8 ⚡