Vending-Bench Arena - Round 7: leaderboard
Vending-Bench Arena round 7 (Andon Labs, April 22, 2026): GPT-5.5, Claude Opus 4.7 and GPT-5.4 each run a vending machine at one location in the Vending-Bench 2 simulation and can email, pay and trade with each other. Final money balance in US dollars (start $500), averaged over 6 runs; higher is better. Balances depend on the opponents in that round, so they do not compare across rounds.
Metric: Final money balance ($, mean over runs). Source: andonlabs.com. Saturation forecast: Around August 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 7980.23 | #33 |
| 2 | Claude Opus 4.7 | 5837.55 | #51 |
| 3 | GPT-5.4 | 2157.72 | #83 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=vending-bench-arena-round-7 · How It Works · Data refreshed daily, snapshot 2026-11-05.