CalBench - Varied Costs - Excess Cost: leaderboard
Metric: Excess displacement cost (cost units). Source: arxiv.org. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 4 Maverick Instruct | 0.24 |
| 2 | Gemini 3.1 Pro (Preview) | 0.36 |
| 3 | DeepSeek V4 Pro | 0.38 |
| 4 | Claude Sonnet 4.6 | 0.43 |
| 5 | Qwen 3.6 Plus | 0.53 |
| 6 | Gemini 3 Flash (Preview) | 0.59 |
| 7 | GPT-5.4 Mini | 1.49 |
Interactive version: theaggregate.ai/benchmark?slug=calbench-varied-costs-excess-cost · How It Works · Data refreshed daily, snapshot 2026-09-19.