Vending-Bench 2

Running a simulated vending business for a year; final money balance. Tests long-horizon coherence.

CategoryAgents
MetricFinal balance (USD)
SourceEpoch AI (evaluator)
LicenseCC BY 4.0
Benchmark pageandonlabs.com
StatusScored
Weight0.9 within Agents
FitDiscrimination 1.52, noise 1.20

Results

#ModelProviderResultDate
1 Gemini 4 Carbon Google $16,484 2026-11-11
2 GPT-6 Astra OpenAI $15,515 2026-11-11
3 GPT-6 Sol OpenAI $14,428 2026-11-11
4 Gemini 4 Argon Google $13,718 2026-11-11
5 GPT-6.1 Sol OpenAI $12,861 2026-11-11
6 Claude Opus 5 Anthropic $11,182 2026-11-11
7 Claude Sonnet 5.5 Anthropic $11,165 2026-11-11
8 Claude Opus 4.7 Anthropic $10,937 2026-11-11
9 Grok 4.7 xAI $10,537 2026-11-11
10 GPT-5.6 Sol OpenAI $9,619 2026-11-11
11 Muse Wonder Meta $9,432 2026-11-11
12 Claude Opus 5.5 Anthropic $9,235 2026-11-11
13 GPT-6.1 Terra OpenAI $9,107 2026-11-11
14 Grok 4.6 xAI $9,047 2026-11-11
15 GLM-5.2 Zhipu AI $8,314 2026-11-11
16 GLM-5.3 Zhipu AI $8,164 2026-11-11
17 Claude Opus 4.6 Anthropic $8,018 2026-11-11
18 GPT-5.5 OpenAI $7,524 2026-11-11
19 GPT-5.6 Terra OpenAI $7,343 2026-11-11
20 Claude Sonnet 4.6 Anthropic $7,204 2026-11-11
21 Muse Spark 1.1 Meta $6,520 2026-11-11
22 Claude Sonnet 5 Anthropic $6,378 2026-11-11
23 Kimi K2.6 Moonshot AI $6,205 2026-11-11
24 GPT-5.4 OpenAI $6,144 2026-11-11
25 GPT-5.3 Codex OpenAI $5,940 2026-11-11
26 Claude Opus 4.8 Anthropic $5,787 2026-11-11
27 Claude Fable 5 High Anthropic $5,680 2026-11-11
28 GLM-5.1 Zhipu AI $5,634 2026-11-11
29 Gemini 3 Pro Preview Google $5,478 2026-11-11
30 Claude Fable 5.1 Anthropic $5,422 2026-11-11
31 Gemini 3.5 Flash Google $5,396 2026-11-11
32 Kimi K3 Moonshot AI $5,165 2026-11-11
33 Qwen3.6 Plus Alibaba $5,115 2026-11-11
34 Gemini 3.8 Flash Google $5,094 2026-11-11
35 Kimi K2.7 Code Moonshot AI $5,083 2026-11-11
36 Claude Fable 5 Low Anthropic $5,019 2026-11-11
37 Claude Opus 4.5 Anthropic $4,967 2026-11-11
38 Claude Fable 5 Max Anthropic $4,967 2026-11-11
39 Grok 4.20 (Reasoning) xAI $4,663 2026-11-11
40 Claude Fable 5 Anthropic $4,530 2026-11-11
41 GLM-5 Zhipu AI $4,432 2026-11-11
42 Claude Fable 5 Medium Anthropic $4,340 2026-11-11
43 Qwen3.6 Max Preview Alibaba $4,254 2026-11-11
44 GPT-5.6 Luna OpenAI $4,095 2026-11-11
45 Grok 4.5 xAI $3,887 2026-11-11
46 Claude Sonnet 4.5 Anthropic $3,839 2026-11-11
47 Gemini 3.1 Pro Preview Custom Tools Google $3,774 2026-11-11
48 Gemini 3 Flash Preview Google $3,635 2026-11-11
49 Claude Haiku 5.5 Anthropic $3,602 2026-11-11
50 GPT-5.2 OpenAI $3,591 2026-11-11
51 DeepSeek V4 Pro DeepSeek $3,285 2026-11-11
52 Claude Opus 4.8 Max Anthropic $2,992 2026-11-11
53 GLM-4.7 Zhipu AI $2,377 2026-11-11
54 MiniMax-M3 MiniMax $2,158 2026-11-11
55 GPT-5.1 OpenAI $1,473 2026-11-11
56 Kimi K2.5 Moonshot AI $1,198 2026-11-11
57 Grok 4.1 Fast (Reasoning) xAI $1,107 2026-11-11
58 DeepSeek-V3.2-Exp DeepSeek $1,034 2026-11-11
59 Gemini 3.1 Pro Preview Google $911 2026-11-11
60 Gemini 2.5 Pro Google $574 2026-11-11
61 Gemini 2.5 Flash Google $549 2026-11-11
62 Qwen3.5 Flash Alibaba $463 2026-11-11
63 Claude Haiku 4.5 Anthropic $459 2026-11-11
64 Qwen3.5 27B Alibaba $202 2026-11-11
65 MiniMax-M2 MiniMax $161 2026-11-11
66 Qwen3 Max Alibaba $72 2026-11-11
67 Grok 4.3 xAI $35 2026-11-11
68 Qwen3.5 Plus Alibaba $1 2026-11-11
69 Qwen3 235B-A22B Thinking 2507 Alibaba $-11 2026-11-11
70 GPT OSS 120B OpenAI $-22 2026-11-11
71 MiniMax-M2.5 MiniMax $-23 2026-11-11
72 GPT-5 Mini OpenAI $-31 2026-11-11