tweetindex

Modal @modal · 27 Aug 2026

Terminal-Bench set the standard for evaluating long-horizon agents. Terminal-Bench Science brings that same high standard to scientific workflows.
6 296Views
46Likes
3Reposts
3Replies
0Quotes
7Bookmarks

Is that a lot?

0.17×vs this author's median36 967 views is typical
0Percentile for this authorof 6 recent posts
2.29×vs 10K–100K median2 751 views is typical
17.63%Reachviews ÷ followers
0.83%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →