Vending-Bench Arena - Round 9: leaderboard
Vending-Bench Arena round 9 (Andon Labs, June 9, 2026): Claude Fable 5 (max reasoning), Claude Opus 4.8 and GPT-5.5 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 $500), averaged over 5 runs; higher is better. Balances depend on the opponents in that round, so they do not compare across rounds.
Metric: Final money balance ($, mean over runs). Source: andonlabs.com. Saturation forecast: Around September 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 8287.91 | #33 |
| 2 | Claude Opus 4.8 | 6195.64 | #39 |
| 3 | Claude Fable 5 (Max) | 4244.26 | #20 (Claude Fable 5) |
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-9 · How It Works · Data refreshed daily, snapshot 2026-11-05.