tweetindex

Sam Earle @Smearle_RH · 29 Jul 2026

In this highly cubular work led by @BigTimeMothy, we harvest many many chunks of (procedural and human-authored) Minecraft worlds, and train diffusion models on them. We can then use these models for high-level-controllable world generation. We can paint biomes in broad strokes
4 105Views
22Likes
2Reposts
1Replies
1Quotes
8Bookmarks

Is that a lot?

1.00×vs this author's median4 105 views is typical
40Percentile for this authorof 5 recent posts
1.82×vs 10K–100K median2 259 views is typical
14.49%Reachviews ÷ followers
0.63%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →