Stanford AI Lab @StanfordAILab · 03 Sep 2026
The most capable models available in the world are using Terminal-Bench-Science to show it (just one week after the benchmark’s release). Congratulations to @StevenDillmann and team for their remarkable impact!
8 615Views
44Likes
5Reposts
2Replies
0Quotes
16Bookmarks
Is that a lot?
0.67×vs this author's median12 882 views is typical
0Percentile for this authorof 6 recent posts
1.96×vs 100K–1M median4 404 views is typical
3.23%Reachviews ÷ followers
0.59%Engagement rateof viewers reacted