Atlas

Benchmarks

← All benchmarks

Excel Modeling Benchmark Template

Professional Work · 2026-07-01

Template-mode score on Excel Modeling Benchmark, where agents complete a stripped template workbook and are graded against the gold workbook.

Top models (higher is better)

ModelScore
Claude Fable 577.0
GPT-5.6 Sol75.3
Claude Opus 574.7
Opus 4.871.1
GPT-5.6 Terra70.3
Sonnet 568.0
Gemini 3.6 Flash64.7
GPT-5.6 Luna64.7
Kimi K364.4
Gemini 3.5 Flash62.5
GPT-5.561.8
GLM-5.255.8
Kimi K2.655.7
Sonnet 4.655.2
MiMo-V2.5-Pro52.1
Qwen3.7-Max51.8
Muse Spark 1.149.4
Gemini 3.1 Pro Preview47.2
DeepSeek-V4-Pro46.1
GPT-5.4 Nano45.1
Qwen3.7-Plus41.5
Gemini 3.5 Flash-Lite40.9
Grok 4.538.3
GPT-5.4 Mini36.0
MiniMax M334.5
Inkling32.5
Inkling-Small18.2
Gemini 3.1 Flash-Lite Preview8.1
Grok 4.36.8
Loading Atlas data…