Today for AI
HOT RADAR
agentHEAT 10.6°

AI CURATED HOT STORY · 11/14/2026

Deepnote Benchmarks 12 AI Models as Slack Teammates, OpenAI Tops Cost-Efficiency

1 reports archived1 independent sourcesupdated 11/14/2026, 17:06:00
Synthesis & Latest Updates
1 Sources Cross-Validated

Deepnote founder Jakub Jurovych released the Agent Initiative Benchmark, placing 12 frontier models into a shared corporate Slack channel as AI teammates to test whether they can judge when to intervene and when to stay silent. OpenAI took the top two spots: GPT-6 Astra scored 8.97 and GPT-6.1 Sol 8.65 at just $0.11 per task, while third-place Anthropic Fable 5.2 scored 8.17 at $1.87 per task, roughly 17x the cost for a lower score and about 16x higher monthly billing for a 70-person company. The widest gap: Fable 5.1 at $3.02 per task versus GPT-6 Luna at $0.007, a 430x price difference for only 1.1 points.

LATEST/Deepnote founder Jakub Jurovych released the Agent Initiative Benchmark, placing 12 frontier models into a shared corporate Slack channel as AI teammates to test whether they can judge when to intervene and when to stay silent. OpenAI took the top two spots: GPT-6 Astra scored 8.97 and GPT-6.1 Sol 8.65 at just $0.11 per task, while third-place Anthropic Fable 5.2 scored 8.17 at $1.87 per task, roughly 17x the cost for a lower score and about 16x higher monthly billing for a 70-person company. The widest gap: Fable 5.1 at $3.02 per task versus GPT-6 Luna at $0.007, a 430x price difference for only 1.1 points.

Hourly heat curve

6 fully observed hours
Now
10.6
Peak
11.818:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. 新智元 (微信公众号)T2·72 pts

    Deepnote Benchmarks 12 AI Models as Slack Teammates, OpenAI Tops Cost-Efficiency

    • The benchmark measures proactive intervention control, i.e. whether a model knows when to speak up and when to stay silent in a group chat, not answer accuracy
    • OpenAI's GPT-6 Astra and GPT-6.1 Sol took the top two spots, with Sol at $0.11 per task, far cheaper than peers at similar scores
    • Anthropic Fable 5.2 cost $1.87 per task for a score of 8.17, a clear efficiency lag; the widest gap reached 430x price for a 1.1-point difference