tweetindex

Ziwen @ziwenxu_ · 03 Sep 2026

Those benchmarks can't be real. GPT-6-Astra is mocking every models on the ground.. It's the closest AGI we have so far. https://t.co/NPm6loKwwe
32 042Views
134Likes
2Reposts
29Replies
4Quotes
0Bookmarks

Is that a lot?

2.75×vs this author's median11 662 views is typical
70Percentile for this authorof 10 recent posts
13.4×vs 10K–100K median2 392 views is typical
180.49%Reachviews ÷ followers
0.53%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →