Benchmark
Vals MGSM
Math benchmarks measure how well models solve maths problems. Scores are percentages.
- Models measured
- 62
- Scores recorded
- 67
- Sources
- 1
- IQ weight
- 2.55
Score distribution
- 88.8%
- Mean
- 90.8%
- Median
- 7.4%
- Std dev
- 85.7%
- 10th percentile
- 93.3%
- 90th percentile
- 41.7%
- Min
This benchmark has a weight of 2.55 for chat models, measured over 62 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 | Llama 4 Scout· Vals AI | 88% |
| 52 | MiniMax M2.1· Vals AI | 87.9% |
| 53 | GPT-4.1 mini· Vals AI | 87.8% |
| 54 | GPT 4.1· Vals AI | 87.7% |
| 55 | Mistral Large2411· Vals AI | 87.2% |
| 56 | gemini-1.5-flash-002· Vals AI | 86.6% |
| 57 | Magistral Small 2509· Vals AI | 86.3% |
| 58 | GPT-4o mini· Vals AI | 86.2% |
| 59 | Grok 2· Vals AI | 86.1% |
| 60 | Command A 03 2025· Vals AI | 85.7% |
| 61 | Mistral Large 3· Vals AI | 85.4% |
| 62 | Claude Haiku 3.5· Vals AI | 84.6% |
| 63 | Mistral Small· Vals AI | 84.2% |
| 64 | Magistral Medium· Vals AI | 74.6% |
| 65 | jamba-1.6-large· Vals AI | 71.2% |
| 66 | GPT-4.1 nano· Vals AI | 69.3% |
| 67 | jamba-1.6-mini· Vals AI | 41.7% |
Showing 51–67 of 67 results.