Gemma 3 is out! Our new models are very efficient, with the biggest 27B model running on just one H100 GPU. You’d need at least 10 times the power to get the same performance from other models.
Gemma 3 is here! Our new open models are incredibly efficient - the largest 27B model runs on just one H100 GPU. You'd need at least 10x the compute to get similar performance from other models ⬇️
Read more:
Missing some Tweet in this thread? You can try to Update
Have you tried our Twitter bot?
You can now unroll any thread without leaving Twitter/X. Here's how to use our Twitter bot to do it.
