Benchmark Almanac

3DGen-Bench

MVDream 1,296 tops its leaderboard
Not in model cards

People compare generated 3D objects in pairs. The authors use their judgments to evaluate automated scoring methods for text-to-3D and image-to-3D generation.

Made by
Yuhan Zhang and colleagues
Released
March 2025
Kind
Image generation and editing
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Measures
Preference
Language
English
In and out
Text, Images → 3D
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
20
Entries on its own leaderboard, 0 models we track

MVDream tops the leaderboard with 1,296

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

Bars start at 600.

  1. 1 MVDream Text-to-3D · Anony. Arena 1,296
  2. 2 InstantMesh Image-to-3D · Anony. Arena 1,271
  3. 3 Stable Zero123 Image-to-3D · Full Arena 1,197
  4. 4 OpenLRM Image-to-3D · Full Arena 1,195
  5. 5 Zero123-XL Image-to-3D · Anony. Arena 1,143
  6. 6 Wonder3D Image-to-3D · Full Arena 1,135
  7. 7 LGM Image-to-3D · Anony. Arena 1,128
  8. 8 LucidDreamer Text-to-3D · Anony. Arena 1,121
  9. 9 Magic123 Image-to-3D · Anony. Arena 1,096
  10. 10 GRM Text-to-3D · Anony. Arena 1,096
  11. 11 Magic3D Text-to-3D · Full Arena 1,018
  12. 12 Latent-NeRF Text-to-3D · Full Arena 959
  13. 13 SyncDreamer Image-to-3D · Anony. Arena 921
  14. 14 DreamFusion Text-to-3D · Full Arena 918
  15. 15 Shap-E Text-to-3D · Full Arena 909
  16. 16 SJC Text-to-3D · Anony. Arena 878
  17. 17 TriplaneGaussian Image-to-3D · Anony. Arena 857
  18. 18 Point-E Text-to-3D · Anony. Arena 812
  19. 19 EscherNet Image-to-3D · Full Arena 735
  20. 20 Free3D Image-to-3D · Anony. Arena 660

Source: Official 3DGen-Bench leaderboard

#EntryScoreSettingDateSource
1 MVDream 1,296 Text-to-3D · Anony. Arena Mar 2025 Official leaderboard
2 InstantMesh 1,271 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
3 Stable Zero123 1,197 Image-to-3D · Full Arena Feb 2025 Official leaderboard
4 OpenLRM 1,195 Image-to-3D · Full Arena Feb 2025 Official leaderboard
5 Zero123-XL 1,143 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
6 Wonder3D 1,135 Image-to-3D · Full Arena Feb 2025 Official leaderboard
7 LGM 1,128 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
8 LucidDreamer 1,121 Text-to-3D · Anony. Arena Mar 2025 Official leaderboard
9 Magic123 1,096 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
10 GRM 1,096 Text-to-3D · Anony. Arena Mar 2025 Official leaderboard
11 Magic3D 1,018 Text-to-3D · Full Arena Mar 2025 Official leaderboard
12 Latent-NeRF 959 Text-to-3D · Full Arena Mar 2025 Official leaderboard
13 SyncDreamer 921 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
14 DreamFusion 918 Text-to-3D · Full Arena Mar 2025 Official leaderboard
15 Shap-E 909 Text-to-3D · Full Arena Mar 2025 Official leaderboard
16 SJC 878 Text-to-3D · Anony. Arena Mar 2025 Official leaderboard
17 TriplaneGaussian 857 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard
18 Point-E 812 Text-to-3D · Anony. Arena Mar 2025 Official leaderboard
19 EscherNet 735 Image-to-3D · Full Arena Feb 2025 Official leaderboard
20 Free3D 660 Image-to-3D · Anony. Arena Feb 2025 Official leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (20 entries).
  • Pass Models scored 20 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on huggingface.co: 20 entries at the last read, none a model we track.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check