Name: 16 parallel runs of Gemma 4 26B A4B on a single NVIDIA DGX Spark! Pushing 18 tok/s per instance and
Uploaded: 2026-06-23T16:09:26+00:00
Description: 16 parallel runs of Gemma 4 26B A4B on a single NVIDIA DGX Spark! Pushing 18 tok/s per instance and a 300 tok/s aggregate. It can even hit 32 parallel runs. This level of concurrency highlights how efficient the architecture is.

Google Gemma

@googlegemma

The official home of Google's Gemma. Lightweight, state-of-the-art open models by Google DeepMind, built on Gemini tech. What will you build? 🚀💻

加入 April 2025

0 正在关注 87.9K 粉丝

Google Gemma@googlegemma

2026.06.23 16:09

16 parallel runs of Gemma 4 26B A4B on a single NVIDIA DGX Spark! Pushing 18 tok/s per instance and a 300 tok/s aggregate. It can even hit 32 parallel runs. This level of concurrency highlights how efficient the architecture is.

显示更多