Benchmark
Vals CorpFin v2
Scores are percentages.
Benchmark sourceScores from Vals AI.
- Models measured
- 120
- Scores recorded
- 125
- Sources
- 1
- IQ weight
- 2.19
Score distribution
- 58.3%
- Mean
- 60.8%
- Median
- 9.3%
- Std dev
- 44.2%
- 10th percentile
- 67.4%
- 90th percentile
- 28.6%
- Min
This benchmark has a weight of 2.19 for chat models, measured over 120 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 |
|---|---|---|
| 101 | Claude Haiku 3.5· Vals AI | 50.8% |
| 102 | Mistral Medium 3· Vals AI | 50.7% |
| 103 | Kimi K2 Instruct· Vals AI | 50.4% |
| 104 | Llama 4 Maverick· Vals AI | 49.7% |
| 105 | Magistral Medium· Vals AI | 47.4% |
| 106 | Llama 4 Scout· Vals AI | 46.8% |
| 107 | GLM 4.7· Vals AI | 46.4% |
| 108 | Command A 03 2025· Vals AI | 46% |
| 109 | GPT 4o· Vals AI | 45.9% |
| 110 | GPT-4o mini· Vals AI | 45.5% |
| 111 | o3 Mini· Vals AI | 45.3% |
| 112 | Mistral Small· Vals AI | 44.2% |
| 113 | Magistral Small 2509· Vals AI | 44% |
| 114 | gemini-2.0-pro-exp-02-05· Vals AI | 43.4% |
| 115 | GPT-4.1 nano· Vals AI | 42.1% |
| 116 | jamba-1.6-large· Vals AI | 41.5% |
| 117 | gemini-1.5-pro-002· Vals AI | 40.5% |
| 118 | Jamba 1.5 Large· Vals AI | 39.4% |
| 119 | Meta Llama 3.1 70B Instruct Turbo· Vals AI | 38.9% |
| 120 | gemini-1.5-flash-002· Vals AI | 38.2% |
| 121 | jamba-1.6-mini· Vals AI | 38% |
| 122 | Llama 3.1 8B Instruct Turbo· Vals AI | 37.8% |
| 123 | Jamba 1.5 Mini· Vals AI | 33.9% |
| 124 | Gemini 2.0 Flash· Vals AI | 33.7% |
| 125 | gemini-1.5-flash-001· Vals AI | 28.6% |
Showing 101–125 of 125 results.