Atlas

Benchmarks

← All benchmarks

FrontierFinance v3.1 — Earnings and Events

Professional Work · 2026-07-06

Earnings and Events use-case slice of Samaya AI's FrontierFinance v3.1, containing 36 of the 220 public queries. It reports rubric qualification rate macro-averaged across queries in the slice.

Top models (higher is better)

ModelScore
Claude Fable 563.9
GPT-5.6 Sol58.0
Opus 4.857.7
DeepSeek-V4-Pro54.5
GPT-5.553.7
GLM-5.252.0
Kimi K2.650.4
Gemini 3.1 Pro Preview42.5
Loading Atlas data…