FrontierFinance v3.1 — Financial Data and Modeling
Professional Work · 2026-07-06
Financial Data and Modeling use-case slice of Samaya AI's FrontierFinance v3.1, containing 70 of the 220 public queries. It reports rubric qualification rate macro-averaged across queries in the slice.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5.1 | 60.9 |
| Gemini 3.6 Flash | 56.2 |
| Gemini 3.8 Flash | 56.2 |
| Claude Fable 5 | 55.6 |
| Opus 4.8 | 51.0 |
| GPT-5.6 Sol | 49.0 |
| GPT-5.5 | 47.6 |
| GLM-5.2 | 47.5 |
| Ling 3.0 Flash Fin | 46.8 |
| Kimi K3 | 46.5 |
| DeepSeek-V4-Pro | 45.3 |
| Gemini 3.1 Pro Preview | 31.4 |
| Kimi K2.6 | 28.8 |