cd /news/ai-infrastructure/google-tpu-and-gpu-performance-mlper… · home › topics › ai-infrastructure › article
[ARTICLE · art-170769] src=blog.google ↗ pub= topic=ai-infrastructure verified=true sentiment=↑ positive

Google TPU and GPU performance: MLPerf v6.1 results

Google Cloud published 10 MLPerf Training v6.1 results spanning seventh-generation Ironwood TPUs and NVIDIA Grace Blackwell GPUs, including the first MLPerf benchmark for agentic reinforcement learning. On Ironwood, 8,192 TPUs reached the target model's benchmark quality in 1.70 minutes, scaling to 4,096 chips at 96% efficiency, while 640 Ironwood TPUs hit the coding benchmark's target accuracy of 0.765 against a 0.69 goal in 18.02 minutes with training chips idle under 3% of the time. On Google Cloud A4X Max Instances, 256 NVIDIA GB300 GPUs reached target quality in 31.96 minutes, alongside five A4X and A4X Max submissions on the agentic RL benchmark.

read1 min views2 publishedNov 16, 2026
Google TPU and GPU performance: MLPerf v6.1 results
Image: Google AI Blog

In the latest MLPerf Training v6.1 benchmark round <sup>1</sup> , Google Cloud published 10 results across seventh-generation Ironwood TPUs and NVIDIA Grace Blackwell GPUs, spanning frontier pre-training and the first MLPerf benchmark for agentic reinforcement learning (RL). Built on Google Cloud’s AI Hypercomputer architecture, these results reflect co-design across silicon, networking, and open software:

  • Pre-training on Ironwood: 8,192 Ironwood TPUs reached the target model's benchmark target quality in 1.70 minutes, scaling up to 4,096 chips with 96% efficiency.
  • Agentic RL on Ironwood: 640 Ironwood TPUs reached the target model’s coding benchmark target accuracy in 18.02 minutes (0.765 vs. the 0.69 goal), running code generation and learning simultaneously so training chips waited under 3% of the time.
  • NVIDIA Grace Blackwell GPUs: On Google Cloud A4X Max Instances, 256 NVIDIA GB300 GPUs reached the target model's benchmark quality in 31.96 minutes, complemented by five A4X and A4X Max submissions on the agentic RL benchmark.

For more details on the MLPerf Training v6.1 results, click here.

── more in #ai-infrastructure 4 stories · sorted by recency
── more on @google cloud 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/google-tpu-and-gpu-p…] indexed:0 read:1min 2026-11-16 · —