Skip to content

Benchmark

Furniture Assembly

Scores are percentages.

Models measured
40
Scores recorded
40
Sources
1
IQ weight
2.20

Score distribution

49.3%
Mean
44.2%
Median
22.3%
Std dev
22.5%
10th percentile
80%
90th percentile
20%
Min
20%90%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.20 for chat models, measured over 40 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5.2max· Epoch AI Benchmarking Hub data90%
2Gemini 4 Neonxhigh· Epoch AI Benchmarking Hub data89.2%
3GPT-6.2 Astramax· Epoch AI Benchmarking Hub data89.2%
4Claude Opus 5.5max· Epoch AI Benchmarking Hub data83.3%
5GPT-6 Astramax· Epoch AI Benchmarking Hub data80%
6GPT-6.1 Solmax· Epoch AI Benchmarking Hub data80%
7GPT-6.1 Terramax· Epoch AI Benchmarking Hub data79.2%
8Gemini 4 Carbonhigh· Epoch AI Benchmarking Hub data77.5%
9Claude Sonnet 5.5max· Epoch AI Benchmarking Hub data75%
10Gemini 4 Argonhigh· Epoch AI Benchmarking Hub data72.5%
11Claude Fable 5.1max· Epoch AI Benchmarking Hub data70%
12Claude Opus 5max· Epoch AI Benchmarking Hub data60.8%
13GPT-6 Solmax· Epoch AI Benchmarking Hub data58.3%
14Gemini 4 Hydrogenxhigh· Epoch AI Benchmarking Hub data57.5%
15GPT-5.6 Solmax· Epoch AI Benchmarking Hub data56.7%
16GPT-5.6 Terramax· Epoch AI Benchmarking Hub data54.2%
17Muse Wonder 1.0max· Epoch AI Benchmarking Hub data54.2%
18Claude Haiku 5.5max· Epoch AI Benchmarking Hub data47.5%
19GPT 5.5xhigh· Epoch AI Benchmarking Hub data44.2%
20GPT-6 Lunamax· Epoch AI Benchmarking Hub data44.2%
21Claude Opus 4.8max· Epoch AI Benchmarking Hub data42.5%
22GPT-5.6 Lunamax· Epoch AI Benchmarking Hub data42.5%
23Grok 4.6xhigh· Epoch AI Benchmarking Hub data40%
24GPT 5.2xhigh· Epoch AI Benchmarking Hub data38.3%
25GPT 5.4xhigh· Epoch AI Benchmarking Hub data37.5%
26Claude Fable 5max· Epoch AI Benchmarking Hub data35.8%
27DeepSeek V4.1 Flashmax· Epoch AI Benchmarking Hub data34.2%
28Kimi K3max· Epoch AI Benchmarking Hub data34.2%
29Claude Opus 4.7max· Epoch AI Benchmarking Hub data33.3%
30Gemini 3.8 Flashhigh· Epoch AI Benchmarking Hub data31.7%
31Claude Opus 4.564k· Epoch AI Benchmarking Hub data28.3%
32Claude Opus 4.6max· Epoch AI Benchmarking Hub data28.3%
33Gemini 3.1 Pro Previewhigh· Epoch AI Benchmarking Hub data26.7%
34Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data26.7%
35Gemini 3.6 Flashhigh· Epoch AI Benchmarking Hub data23.3%
36Grok 4.5high· Epoch AI Benchmarking Hub data22.5%
37Kimi K2.6· Epoch AI Benchmarking Hub data21.7%
38Grok 4.7xhigh· Epoch AI Benchmarking Hub data20.8%
39Mistral Large 4high· Epoch AI Benchmarking Hub data20.8%
40Qwen3.8 Maxxhigh· Epoch AI Benchmarking Hub data20%

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)