Skip to content

Benchmark

BlenderWorld · Image/video-input overall score

Higher scores are better.

Models and systems measured
13
Scores recorded
13
Sources
1
IQ weight
1.38

Score distribution

55.8
Mean
56.1
Median
13.5
Std dev
33.2
10th percentile
72.1
90th percentile
30.6
Min
30.680.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.38 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1GPT-6 Astrabenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs80.7
2Claude Opus 5benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs72.1
3DeepSeek V4.1 Flashbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs65.5
4GPT-5.6 Solbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs65.5
5Qwen3.8 Maxbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.1
6GLM 5.3 Flashbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs57.8
7Claude Opus 4.8benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs56.1
8Kimi K3benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs49.4
9Qwen 3.8 27Bbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48.6
10Gemini 3.8 Flashbenchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs33.2
11MiniMax M3benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs30.6

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Grok 4.6benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs55.4
Seed2.1 Pro 0915benchmark-owner paper run · 145 image or video reference tasks · Blender 4.2.10 / Blender MCP paper setup · GPT-5.6 Terra main judge · generation-agent harness and reasoning effort not stated on public leaderboard· Reported by BlenderWorld authors / Alibaba Token Hub· via BenchBoard docs49.5

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.