Ofir Press @OfirPress · 05 May 2026
1) Our team at Meta has a tough new coding benchmark challenging models to code entire programs including ffmpeg and the PHP compiler from scratch. 2) Top accuracy is 0% 3) We will be making the benchmark harder.
142 146Views
1 029Likes
62Reposts
37Replies
6Quotes
251Bookmarks
Is that a lot?
44.3×vs this author's median3 207 views is typical
91Percentile for this authorof 11 recent posts
39.5×vs 10K–100K median3 598 views is typical
7.3× audienceReachviews ÷ followers
0.80%Engagement rateof viewers reacted