Atlas

Benchmarks

← All benchmarks

FrontierFinance v3.1 — Qualitative and Contextual Information Rubrics

Professional Work · 2026-07-06

Qualitative and Contextual Information rubric-category slice in the official FrontierFinance v3.1 system breakdown, with a reported denominator of 1,040 rubric criteria. It reports micro-averaged rubric qualification rate within that displayed category.

Top models (higher is better)

ModelScore
Claude Fable 5.147.3
GPT-5.6 Sol44.2
Gemini 3.8 Flash43.2
Kimi K342.6
Claude Fable 541.2
GPT-5.540.4
Ling 3.0 Flash Fin39.4
GLM-5.238.9
Opus 4.836.7
DeepSeek-V4-Pro36.5
Gemini 3.6 Flash36.5
Kimi K2.628.6
Gemini 3.1 Pro Preview25.7
Loading Atlas data…