tweetindex

Lyr Aura @Lyr_Soul · 04 Sep 2026

The thing I keep coming back to with K2 Horizon is the training record. We normally see the finished model and a few evaluation charts. Here, IFM is giving researchers checkpoints, logs, training code and evaluations too. That changes what “open” can mean in practice.
10 859Views
109Likes
94Reposts
14Replies
0Quotes
0Bookmarks

Is that a lot?

1.00×vs this author's median10 859 views is typical
44Percentile for this authorof 9 recent posts
5.57×vs 10K–100K median1 948 views is typical
48.13%Reachviews ÷ followers
2.00%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →