"Spent a lot on my rig but I can only run Gemma 4"
Summary
Google's first data center satellite, equipped with four Tensor Processing Units generating up to 1 kW, is now online in orbit to test large inference workloads in space, though TPU power limits restrict it to running a single chip for about 20 minutes at a time.
Similar Articles
Google’s first data center satellite is online and writing poetry
Google's Suncatcher prototype space data center satellite is online and communicating with the ground after launching on a SpaceX Falcon 9, running Google DeepMind's Gemma 4 on onboard TPUs to perform inference in orbit. The test will assess how the chips handle space radiation and thermal stress, with two more satellites planned for 2027.
@Saboo_Shubham_: This is INSANE. A huge step for humanity. Google Gemma 4 is a actually running at a Data Center in Space. Read that aga…
Google's Project Suncatcher reportedly ran a Gemma model on a data center in space, marking a notable milestone for orbital AI compute infrastructure.
@Google: Our TPUs just phoned home... from space. In early October, we sent four TPUs into orbit to test whether our AI hardware…
Google reported successfully operating four of its TPUs in orbit, confirming the AI hardware could power on, process a prompt, and transmit a response back to Earth after being launched into space in early October.
Gemma 12b less than 10 watts 6.5pp 1.3tg
Running Gemma 12B model on a Google Pixel 10 Pro using llama.cpp achieves 6.5 tokens per second prompt processing and 1.3 tokens per second generation with under 10 watts power consumption, demonstrating efficient on-device AI inference.
New Google Gemma 4 12B Claims Near-26B Performance - We Tested Both!
Google's new Gemma 4 12B model claims near-26B performance. In a local test on RTX 4090, the 26B-A4B model was faster and better but the 12B used less VRAM, making it suitable for laptops.