Benchmark Almanac

Code4Scene

GPT-6 Astra 0.62 leads since September 2026
Not in model cards

Agents build and edit Unreal Engine scenes. The evaluator inspects the scene itself for correct edits, intact assets, and physical validity.

Made by
Xiaokang Ye and colleagues
Released
September 2026
Kind
Image generation and editing
Level
University
Scored
Checked automatically
Answers
Code
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text, Images, 3D → Code, Actions, 3D
Interaction
Agentic
Environment
Unreal Engine
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
12
Models scored on its own leaderboard

GPT-6 Astra leads with 0.62 among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-6 Astra max 0.62
  2. 2 Gemini 3.8 Flash high 0.62
  3. 3 Claude Fable 5.1 max 0.61
  4. 4 Claude Opus 5 max 0.59
  5. 5 GPT-5.6 Sol high 0.55
  6. 6 Muse Spark 1.3 medium 0.50
  7. 7 GLM-5.3-Flash max 0.41
  8. 8 Qwen 3.8 27B thinking off 0.38
  9. 9 Grok 4.6 high 0.38
  10. 10 Inkling high 0.31
  11. 11 Gemma 4 31B IT thinking on 0.30
  12. 12 DeepSeek V4.1 Flash high 0.23

Source: Official Code4Scene leaderboard

#ModelScoreSettingDateSource
1 GPT-6 Astra 0.62 max Sep 2026 Official leaderboard
2 Gemini 3.8 Flash 0.62 high Sep 2026 Official leaderboard
3 Claude Fable 5.1 0.61 max Sep 2026 Official leaderboard
4 Claude Opus 5 0.59 max Jul 2026 Official leaderboard
5 GPT-5.6 Sol 0.55 high Jul 2026 Official leaderboard
6 Muse Spark 1.3 0.50 medium Sep 2026 Official leaderboard
7 GLM-5.3-Flash 0.41 max Aug 2026 Official leaderboard
8 Qwen 3.8 27B 0.38 thinking off Aug 2026 Official leaderboard
9 Grok 4.6 0.38 high Aug 2026 Official leaderboard
10 Inkling 0.31 high Jul 2026 Official leaderboard
11 Gemma 4 31B IT 0.30 thinking on Apr 2026 Official leaderboard
12 DeepSeek V4.1 Flash 0.23 high Sep 2026 Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Text-to-Scene, Image-to-Scene and Repair F1. Gemini 3.8 Flash leads 3 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Text-to-Scene

12 models we track have a score here.

  1. 1 Claude Fable 5.1 ↑2 0.79
  2. 2 GPT-6 Astra ↓1 0.72
  3. 3 Claude Opus 5 ↑1 0.72
  4. 4 GPT-5.6 Sol ↑1 0.71
  5. 5 Gemini 3.8 Flash ↓3 0.66
  6. 6 Muse Spark 1.3 0.65
  7. 7 Grok 4.6 ↑2 0.57
  8. 8 Qwen 3.8 27B 0.56
  9. 9 GLM-5.3-Flash ↓2 0.51
  10. 10 Gemma 4 31B IT ↑1 0.47

The best score, 0.62, costs $13 a task

Runs on the cost frontier score higher than any run that's cheaper. Gemini 3.8 Flash at high comes within 0 of the best for $1.92 a task, 7 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Google DeepMind Meta OpenAI Cost frontier One model at each effort level
0.00 0.20 0.40 0.60 0.80 $0.01 $0.10 $1.00 $10 $100 Cost per task, log scale GPT-6 Astra, max Gemini 3.8 Flash, high Muse Spark 1.3, medium Gemma 4 31B IT, thinking on 0.00 0.20 0.40 0.60 0.80 $0.01 $0.10 $1.00 $10 $100 Cost per task, log scale GPT-6 Astra, max Muse Spark 1.3, medium

Source: Official Code4Scene leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-6 Astra max 0.62 $13
Gemini 3.8 Flash high 0.62 $1.92
Claude Fable 5.1 max 0.61 $11
Claude Opus 5 max 0.59 $19
GPT-5.6 Sol high 0.55 $2.98
Muse Spark 1.3 medium 0.50 $0.09
GLM-5.3-Flash max 0.41 $0.40
Qwen 3.8 27B thinking off 0.38 $0.38
Grok 4.6 high 0.38 $2.29
Qwen 3.8 27B thinking on 0.37 $0.30
Inkling high 0.31 $0.48
Gemma 4 31B IT thinking on 0.30 $0.03
Gemma 4 31B IT thinking off 0.28 $0.02
DeepSeek V4.1 Flash high 0.23 $0.12

3 models have held number one on the official leaderboard since its release in September 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic Google DeepMind OpenAI
Own leaderboard Not tested since Sep 2026

Source: Official Code4Scene leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
0.20 0.30 0.40 0.50 0.60 0.70 2026 0.20 0.30 0.40 0.50 0.60 0.70 2026

Source: Official Code4Scene 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 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (12 entries).
  • Pass Models scored 12 models we track.
  • Pass Newest score The newest is from September 2026, for DeepSeek V4.1 Flash.
  • Note Frontier models Only 2 of the top 10 have a score. The strongest tested is GPT-6 Astra (#4).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on simworld-ai-code4scene.static.hf.space: 14 rows for models we track at the last read.
  • 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