Vending Bench 2
AI benchmark results and model performance
Type: numerical
Recorded results
15
Average score
5026
Score range
351 - 16484
Leading model
16484 - Gemini 4 Carbon
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Gemini 4 Carbon | 28 Oct 2026 | 16,484 | Google-reported launch evaluation; Final bank balance (USD); source chart: https://storage.googleapis.com/gweb-uniblog-publish-prod/images/gemini_4_carbon__evals-table.original.png; methodology: deepmind.google/models/evals-methodology/gemini-4-carbon | Yes | Source | |
| Claude Opus 4.6 | 05 Feb 2026 | 8,018 | Final balance | Yes | Source | |
| Claude Sonnet 4.6 | 17 Feb 2026 | 7,204 | Max effort final balance | Yes | Source | |
| GLM 5.1 | 07 Apr 2026 | 5,634 | - | Yes | Source | |
| Gemini 3 Pro Image Preview (Nano Banana Pro) | 20 Nov 2025 | 5,478 | inferred modality/version alias from gemini-3-pro-preview | Yes | Source | |
| Gemini 3 Pro Preview | 18 Nov 2025 | 5,478 | - | Yes | Source | |
| Claude Opus 4.5 | 24 Nov 2025 | 4,967 | 8k Thinking | No | Source | |
| Claude Sonnet 4.5 | 29 Sept 2025 | 3,839 | - | No | Source | |
| Gemini 3 Flash Preview | 17 Dec 2025 | 3,635 | Higher is better, value in dollars | Yes | Source | |
| Grok 4 | 10 Jul 2025 | 1,999 | - | No | Source | |
| GPT 5.1 | 12 Nov 2025 | 1,473 | - | No | Source | |
| Gemini 2.5 Pro Preview (2025-06-05) | 05 Jun 2025 | 574 | - | No | Source | |
| Grok 4.1 Non Thinking | 17 Nov 2025 | 351 | - | No | Source |
