Leaderboard PLANTS

One or more photos reconstructed into a 3D mesh.

Ranked by human votes · 1614 cast · see the AI-judge board →

Share Post on X
How ranking works

Rank groups models whose 95% bootstrap CIs overlap into the same rank (they are not statistically separable). BT = Bradley–Terry score (Elo-scaled); the bar shows the 95% CI with the point estimate marked. Every board ranks a SINGLE method — scores from different methods come from disconnected match pools and aren't comparable, so there is no cross-method ranking. Elo updates live per vote. Recompute in Admin.

Image→3D reconstruction

11 generators
Rank Generator BT score 95% interval BT 345.9–1625.3 Trend Votes Status
1 TRELLIS 2 papercode 1397.9
[1310.2, 1625.3]
5 25 more votes → firm
2 Meshy 6 papercode 1128.5
[1040.5, 1231.8]
64 firm
2 Pixal3D papercode 1031.7
[913.8, 1111.0]
46 firm
2 TRELLIS via Replicate papercode 993.8
[679.8, 1255.1]
10 20 more votes → firm
TRELLIS via fal papercode 955.9
[345.9, 1412.1]
5 25 more votes → firm
2 Hunyuan3D 3.1 papercode 986.6
[856.6, 1092.4]
33 firm
2 Rodin/Hyper3D papercode 981.8
[850.9, 1136.5]
19 11 more votes → firm
2 Hunyuan3D v2 papercode 950.0
[784.4, 1080.5]
16 14 more votes → firm
3 Hunyuan3D v3 papercode 929.6
[790.1, 1027.2]
38 firm
3 SAM 3D papercode 840.7
[671.5, 915.8]
63 firm
3 TripoSR papercode 813.3
[384.8, 980.3]
8 22 more votes → firm
3 InstantMesh papercode 711.9
[480.9, 1022.6]
1 29 more votes → firm

95% credible interval · point estimate · trend · firm a rank backed by 30+ votes; below that, Status counts the votes still needed · click a row for detail

Filters & bias audit

Bias audit — left(A) win rate 0.5 (≈0.50 = unbiased) · tie 0.07 · bad 0.162

VLM judge (Sonnet 4.6, multi-view) — automated LLM-judge rankings by paradigm

Loading automated rankings…