Deepnote Agent Initiative Benchmark: leaderboard

Deepnote tests when a proactive workplace agent should investigate and help, and when it should remain silent. All 12 models face the same 50 synthetic company situations at source-declared high reasoning effort, using 11 service APIs and 180 seconds of model response time (tool time excluded). Two mutually blind LLM evaluators grade against a reference solution and complete action log; the0–10mean covers all50situations. Exact judge models and API snapshots are unreported. Higher is better.

Metric: Mean blind evaluator task score (0-10). Source: deepnote.com. Saturation forecast: Around February 2027. 12 models tracked. Updated 14 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraHigh8.97#6
2GPT-6.1 SolHigh8.65#10
3Claude Fable 5.2High8.17#1
4Claude Opus 5.5High7.84#5
5Claude Fable 5.1High7.5#15
6Claude Sonnet 5.5High6.65#19
7GPT-6 LunaHigh6.38#99
8DeepSeek V4.1 FlashHigh6.35#62
9Mistral Large 4High6.09#137
10Claude Haiku 5.5High6.05#70
11Gemini 4 ArgonHigh5.6#12
12Grok 4.7High4.51#47

No result here: #2 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=deepnote-agent-initiative-benchmark · How It Works · Data refreshed daily, snapshot 2026-11-16.