Benchmark
Vending-Bench 2
Agentic benchmarks measure how well models carry out tasks that take several steps. Higher scores are better.
- Models measured
- 69
- Scores recorded
- 87
- Sources
- 3
- IQ weight
- 2.11
Score distribution
- 6,546.7
- Mean
- 5,165
- Median
- 8,742.6
- Std dev
- 458.9
- 10th percentile
- 11,181.9
- 90th percentile
- -31.2
- Min
This benchmark has a weight of 2.11 for chat models, measured over 69 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6Anthropic Opus 4.6 launch page · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80,017.6 |
| 2 | Gemini 4 Neon· Epoch AI Benchmarking Hub data | 16,843.4 |
| 3 | Gemini 4 Carbon· Epoch AI Benchmarking Hub dataValues from 2 sources
| 16,484.2 |
| 4 | GPT-6 Astra· Epoch AI Benchmarking Hub data | 15,514.7 |
| 5 | Gemini 4 Hydrogen· Epoch AI Benchmarking Hub data | 14,549.5 |
| 6 | GPT-6 Sol· Epoch AI Benchmarking Hub data | 14,427.8 |
| 7 | Gemini 4 Argon· Epoch AI Benchmarking Hub data | 13,718.2 |
| 8 | GPT-6.1 Sol· Epoch AI Benchmarking Hub data | 12,860.6 |
| 9 | Claude Opus 5· Epoch AI Benchmarking Hub data | 11,181.9 |
| 10 | Claude Sonnet 5.5· Epoch AI Benchmarking Hub data | 11,165.3 |
| 11 | Claude Opus 4.7final balance · max effort · benchmark-native context management· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 10,937 |
| 12 | Claude Opus 4.7· Epoch AI Benchmarking Hub data | 10,936.8 |
| 13 | Grok 4.7· Epoch AI Benchmarking Hub data | 10,536.8 |
| 14 | Claude Fable 5.2· Epoch AI Benchmarking Hub data | 9,628.5 |
| 15 | GPT-5.6 Sol· Epoch AI Benchmarking Hub data | 9,619.4 |
| 16 | Muse Wonder 1.0· Epoch AI Benchmarking Hub data | 9,432.2 |
| 17 | Claude Opus 5.5· Epoch AI Benchmarking Hub data | 9,235.2 |
| 18 | GPT-6.1 Terra· Epoch AI Benchmarking Hub data | 9,107.5 |
| 19 | Grok 4.6· Epoch AI Benchmarking Hub data | 9,047 |
| 20 | GLM 5.2· Epoch AI Benchmarking Hub data | 8,313.8 |
| 21 | GLM 5.3· Epoch AI Benchmarking Hub data | 8,163.6 |
| 22 | Claude Opus 4.6· Epoch AI Benchmarking Hub dataValues from 2 sources
| 8,017.6 |
| 23 | Claude Opus 4.7final balance · high effort · benchmark-native context management· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 7,971 |
| 24 | GPT 5.5· Epoch AI Benchmarking Hub data | 7,523.8 |
| 25 | GPT-5.6 Terra· Epoch AI Benchmarking Hub data | 7,343.2 |
| 26 | Claude Sonnet 4.6final balance · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 7,204.1 |
| 27 | Claude Sonnet 4.6· Epoch AI Benchmarking Hub data | 7,204.1 |
| 28 | Claude Sonnet 4.6final balance · high effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 6,625.1 |
| 29 | Muse Spark 1.1· Epoch AI Benchmarking Hub data | 6,520.5 |
| 30 | Claude Sonnet 5· Epoch AI Benchmarking Hub data | 6,377.7 |
| 31 | Kimi K2.6· Epoch AI Benchmarking Hub data | 6,204.6 |
| 32 | GPT 5.4· Epoch AI Benchmarking Hub data | 6,144.2 |
| 33 | GPT-5.3 Codex· Epoch AI Benchmarking Hub data | 5,940.1 |
| 34 | Claude Opus 4.8· Epoch AI Benchmarking Hub data | 5,787.4 |
| 35 | Claude Fable 5high· Epoch AI Benchmarking Hub data | 5,680.3 |
| 36 | GLM 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 5,634.4 |
| 37 | GLM 5.1GLM-5.1 official release table · average net assets· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 5,634.4 |
| 38 | Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 5,478.2 |
| 39 | Gemini 3 Pro PreviewGemini 3 Pro Model Card · results as of November 2025 · mean net worth· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 5,478.2 |
| 40 | Claude Fable 5.1· Epoch AI Benchmarking Hub data | 5,421.6 |
| 41 | Gemini 3.5 Flash· Epoch AI Benchmarking Hub data | 5,396.4 |
| 42 | Kimi K3· Epoch AI Benchmarking Hub data | 5,165 |
| 43 | Qwen3.6 Plus· Epoch AI Benchmarking Hub data | 5,114.9 |
| 44 | Gemini 3.8 Flash· Epoch AI Benchmarking Hub data | 5,093.8 |
| 45 | Kimi K2.7 Code· Epoch AI Benchmarking Hub data | 5,082.9 |
| 46 | Claude Fable 5low· Epoch AI Benchmarking Hub data | 5,018.5 |
| 47 | Claude Opus 4.5final balance · high effort · 8,192 reasoning tokens/turn · benchmark-native context management· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 4,967.1 |
| 48 | Claude Opus 4.5· Epoch AI Benchmarking Hub data | 4,967.1 |
| 49 | Claude Fable 5max· Epoch AI Benchmarking Hub data | 4,966.6 |
| 50 | Grok 4.20· Epoch AI Benchmarking Hub data | 4,662.8 |
Showing 1–50 of 87 results.
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats