Atlas

Benchmarks

Epoch Automation Reports (November 2026) — Data Insights

Human rubric score on two data-insight generation tasks in Epoch's November 17, 2026 report, graded 1-4 per criterion with section weights: Open-ended (average difficulty) and Token prices (difficult), combined with Epoch's 3:2 average-to-difficult weights.

Models
6
Top score
85.0
Leader
Claude Opus 5.5
Updated
Nov 17, 2026
Resolution
—
Saturation
—

No signal or difficulty estimate. Too few models have been measured on this benchmark to estimate its difficulty.

Organizations
6 dated models · 2 frontier modelsClosed weightsOpen weights
Data: Epoch AIatlas.village.dev

Epoch Automation Reports (November 2026) — Data Insights leaderboard

#ModelSettingScoreSource
1Claude Opus 5.5thinking · Epoch Automation Reports: Claude Code; highest available reasoning setting (native value unreported); memory and skill writing disabled
85.0
2Claude Fable 5.2thinking · Epoch Automation Reports: Claude Code; highest available reasoning setting (native value unreported); memory and skill writing disabled
85.0
3Claude Fable 5.1thinking · Epoch Automation Reports: Claude Code; highest available reasoning setting (native value unreported); memory and skill writing disabled
75.0
4Gemini 4 Argonthinking · Epoch Automation Reports: Antigravity; highest available reasoning setting (native value unreported); memory and skill writing disabled
73.0
5GPT-6.2 Astrathinking · Epoch Automation Reports: Codex; highest available reasoning setting (native value unreported); memory and skill writing disabled
69.5
6GPT-6 Astrathinking · Epoch Automation Reports: Codex; highest available reasoning setting (native value unreported); memory and skill writing disabled
69.0

Related benchmarks

Loading Atlas data…