Benchmark
Debate (Lech Mazur)
Benchmark sourceScores from Lech Mazur suite data.
- Models measured
- 64
- Scores recorded
- 67
- Sources
- 1
- IQ weight
- 2.13
Score distribution
- 1,501.3
- Mean
- 1,520.3
- Median
- 139.9
- Std dev
- 1,349.7
- 10th percentile
- 1,676.2
- 90th percentile
- 1,006.3
- Min
This benchmark has a weight of 2.13 for chat models, measured over 64 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 51 | MiniMax M2.7· Lech Mazur suite data | 1,412.1 |
| 52 | Grok 4.20Reasoning· Lech Mazur suite data | 1,389.8 |
| 53 | Gemini 3.5 Flash· Lech Mazur suite data | 1,386.5 |
| 54 | Grok 4.20Non-Reasoning· Lech Mazur suite data | 1,382.3 |
| 55 | ERNIE 5.1· Lech Mazur suite data | 1,369.5 |
| 56 | MiMo V2 Pro· Lech Mazur suite data | 1,368.5 |
| 57 | Qwen3.5 397B A17B· Lech Mazur suite data | 1,368.4 |
| 58 | Hy3 (Deep Infra)high· Lech Mazur suite data | 1,365.6 |
| 59 | Step 3.7 Flashhigh· Lech Mazur suite data | 1,355 |
| 60 | Grok 4.3· Lech Mazur suite data | 1,349.7 |
| 61 | DeepSeek V3.2· Lech Mazur suite data | 1,340.7 |
| 62 | Mistral Medium 3.5high· Lech Mazur suite data | 1,321.4 |
| 63 | Gemini 3.1 Flash Lite· Lech Mazur suite data | 1,308.4 |
| 64 | GPT OSS 120B· Lech Mazur suite data | 1,242 |
| 65 | ERNIE 5.0· Lech Mazur suite data | 1,225.3 |
| 66 | Mistral Large 3· Lech Mazur suite data | 1,169.5 |
| 67 | Llama 4 Maverick· Lech Mazur suite data | 1,006.3 |
Showing 51–67 of 67 results.