tweetindex

Standard Intelligence @si_pbc · 23 Feb 2026

We built eval infrastructure that drives over 1M rollouts per hour across 80,000 forking VMs, each with 1 vCPU and 8GB of RAM; a single H100 can control 42 of these in parallel. These forking VMs are powered by @modal GPUs - the combination of instant spinup and low latency to
43 543Views
297Likes
15Reposts
7Replies
3Quotes
132Bookmarks

Is that a lot?

15.0×vs 10K–100K median2 896 views is typical
4.2× audienceReachviews ÷ followers
0.74%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →