Skip to content

Benchmark

Buyout Game (Lech Mazur)

Models measured
31
Scores recorded
35
Sources
1
IQ weight
2.06

Score distribution

1,502.7
Mean
1,522
Median
230.1
Std dev
1,206.3
10th percentile
1,791.6
90th percentile
997.1
Min
997.11,975.4
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.06 for chat models, measured over 31 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 5.5high reasoning· Lech Mazur suite data1,975.4
2GPT 5.4high reasoning· Lech Mazur suite data1,960.6
3Claude Opus 4.7high reasoning· Lech Mazur suite data1,878.4
4GLM 5· Lech Mazur suite data1,791.6
5Claude Opus 4.6high reasoning· Lech Mazur suite data1,759.7
6DeepSeek V4 Pro· Lech Mazur suite data1,681.6
7Gemini 3.5 Flash· Lech Mazur suite data1,667.2
8Claude Sonnet 4.6high reasoning· Lech Mazur suite data1,656.2
9GPT 5.4no reasoning· Lech Mazur suite data1,654
10Hy3 (Deep Infra)high· Lech Mazur suite data1,620.9
11Gemini 3.1 Flash Lite· Lech Mazur suite data1,615.3
12GLM 5.1· Lech Mazur suite data1,604.9
13Gemini 3.1 Pro Preview· Lech Mazur suite data1,563.8
14DeepSeek V4 Flash· Lech Mazur suite data1,555.8
15Claude Sonnet 4.6no reasoning· Lech Mazur suite data1,549.9
16Grok 4.20Reasoning· Lech Mazur suite data1,549.4
17Grok 4.3· Lech Mazur suite data1,525.1
18Gemma 4 31B ITreasoning· Lech Mazur suite data1,522
19MiMo V2.5 Pro· Lech Mazur suite data1,516.5
20Claude Opus 4.6no reasoning· Lech Mazur suite data1,510.2
21Kimi K2.6· Lech Mazur suite data1,467.6
22Kimi K2.5thinking· Lech Mazur suite data1,461.7
23Qwen3.6 Max Preview· Lech Mazur suite data1,458.7
24MiMo V2 Pro· Lech Mazur suite data1,427.5
25Step 3.5 Flash· Lech Mazur suite data1,415.6
26DeepSeek V3.2· Lech Mazur suite data1,380.6
27MiniMax M2.7· Lech Mazur suite data1,374.2
28Mistral Medium 3.5high· Lech Mazur suite data1,332.2
29Qwen3.5 397B A17B· Lech Mazur suite data1,309.6
30Grok 4.20Non-Reasoning· Lech Mazur suite data1,224.7
31ERNIE 5.1· Lech Mazur suite data1,210.5
32Mistral Large 3· Lech Mazur suite data1,206.3
33Llama 4 Maverick· Lech Mazur suite data1,127.5
34ERNIE 5.0· Lech Mazur suite data1,043
35GPT OSS 120B· Lech Mazur suite data997.1