Atlas

Benchmarks

← All benchmarks

FrontierFinance v3.1 — Format and Presentation Rubrics

Professional Work · 2026-07-06

Format and Presentation rubric-category slice in the official FrontierFinance v3.1 system breakdown, with a reported denominator of 131 rubric criteria. It reports micro-averaged rubric qualification rate within that displayed category.

Top models (higher is better)

ModelScore
GPT-5.6 Sol79.4
GPT-5.577.5
Gemini 3.8 Flash77.1
Claude Fable 5.175.6
Claude Fable 574.4
Gemini 3.6 Flash74.0
Opus 4.873.3
Kimi K2.670.3
GLM-5.267.2
Ling 3.0 Flash Fin66.9
Kimi K365.9
DeepSeek-V4-Pro65.1
Gemini 3.1 Pro Preview53.4
Loading Atlas data…