FrontierFinance v3.1 — Company Research
Professional Work · 2026-07-06
Company Research use-case slice of Samaya AI's FrontierFinance v3.1, containing 32 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 | 53.7 |
| Ling 3.0 Flash Fin | 47.8 |
| Gemini 3.8 Flash | 46.6 |
| Kimi K3 | 45.9 |
| GPT-5.6 Sol | 42.1 |
| Gemini 3.6 Flash | 41.2 |
| Claude Fable 5 | 40.5 |
| GPT-5.5 | 39.7 |
| Opus 4.8 | 38.9 |
| GLM-5.2 | 38.5 |
| Kimi K2.6 | 33.3 |
| DeepSeek-V4-Pro | 32.1 |
| Gemini 3.1 Pro Preview | 29.7 |