CalBench - Varied Costs - Burden Unfairness: leaderboard
Standard deviation of per-agent excess burden; lower means fairer burden allocation. Canonical mixed-agent calendar negotiation, paper v3 Table 1. Uniform and varied displacement costs are separate settings; 45 or 135 evaluated agent-game seats per model, not homogeneous teams. Means retain original units. This series reports Burden unfairness (standard deviation). Source: https://arxiv.org/html/2605.09823v3.
Metric: Burden unfairness (standard deviation). Source: arxiv.org. Saturation forecast: Not forecast. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 4 Maverick Instruct | 0.63 | #467 |
| 2 | Qwen 3.6 Plus | 0.66 | #124 |
| 3 | Gemini 3 Flash (Preview) | 0.72 | #90 |
| 4 | Gemini 3.1 Pro (Preview) | 0.75 | #64 |
| 5 | Claude Sonnet 4.6 | 0.91 | #94 |
| 6 | DeepSeek V4 Pro | 0.96 | #112 |
| 7 | GPT-5.4 Mini | 1.38 | #228 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=calbench-varied-costs-burden-unfairness · How It Works · Data refreshed daily, snapshot 2026-11-05.