Benchmark
Vals Autoform-Hard
Math benchmarks measure how well models solve maths problems. Scores are percentages.
- Models measured
- 8
- Scores recorded
- 8
- Sources
- 1
- IQ weight
- 1.55
Score distribution
- 55.4%
- Mean
- 71.4%
- Median
- 34.3%
- Std dev
- 0%
- 10th percentile
- 85.7%
- 90th percentile
- 0%
- Min
This benchmark has a weight of 1.55 for chat models, measured over 8 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 |
|---|---|---|
| 1 | GPT-6.2 Astra· Vals AI | 87.5% |
| 2 | Claude Fable 5.2· Vals AI | 85.7% |
| 3 | GPT-6.2 Sol· Vals AI | 85.7% |
| 4 | Gemini 4 Carbon· Vals AI | 71.4% |
| 5 | Gemini 4 Hydrogen· Vals AI | 64.3% |
| 6 | Gemini 4 Argon· Vals AI | 48.2% |
| 7 | Beam 1.0 501B A23B· Vals AI | 0% |
| 8 | MiniMax M3.1 Flash· Vals AI | 0% |