FrontierFinance v3.1 — Comparative Analysis Rubrics
Professional Work · 2026-07-06
Comparative Analysis rubric-category slice in the official FrontierFinance v3.1 system breakdown, with a reported denominator of 264 rubric criteria. It reports micro-averaged rubric qualification rate within that displayed category.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5.1 | 54.2 |
| GPT-5.6 Sol | 53.0 |
| Opus 4.8 | 50.4 |
| Claude Fable 5 | 50.2 |
| Kimi K3 | 49.4 |
| Gemini 3.8 Flash | 48.5 |
| GPT-5.5 | 45.6 |
| Gemini 3.6 Flash | 45.5 |
| GLM-5.2 | 45.5 |
| Ling 3.0 Flash Fin | 45.5 |
| Gemini 3.1 Pro Preview | 42.9 |
| DeepSeek-V4-Pro | 39.8 |
| Kimi K2.6 | 35.1 |