Vending-Bench Arena - Round 8: leaderboard
Metric: Final money balance ($, mean over runs). Source: andonlabs.com. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 10626.96 |
| 2 | Claude Opus 4.7 | 5826.74 |
| 3 | Claude Opus 4.8 | 5188.35 |
Interactive version: theaggregate.ai/benchmark?slug=vending-bench-arena-round-8 · How It Works · Data refreshed daily, snapshot 2026-09-19.