SimpleBench
Trick questions about everyday physical and social situations that people find easy.
| Category | Reasoning |
|---|---|
| Metric | Accuracy |
| Source | Epoch AI (evaluator) |
| License | CC BY 4.0 |
| Benchmark page | simple-bench.com |
| Status | Scored |
| Weight | 1.0 within Reasoning |
| Fit | Discrimination 0.76, noise 0.35 |
Results
| # | Model | Provider | Result | Date |
|---|---|---|---|---|
| 1 | Claude Fable 5 Max | Anthropic | 81.9% | 2026-11-11 |
| 2 | Claude Opus 5 | Anthropic | 80.6% | 2026-11-11 |
| 3 | Gemini 3.1 Pro Preview | 79.6% | 2026-11-11 | |
| 4 | GPT-5.5 Pro | OpenAI | 76.9% | 2026-11-11 |
| 5 | GPT-5.5 Pro High | OpenAI | 76.9% | 2026-11-11 |
| 6 | Gemini 3.5 Flash | 76.7% | 2026-11-11 | |
| 7 | Gemini 3 Pro Preview | 76.4% | 2026-11-11 | |
| 8 | Grok 4.6 | xAI | 75.9% | 2026-11-11 |
| 9 | Muse Spark 1.2 | Meta | 74.5% | 2026-11-11 |
| 10 | GPT-5.4 Pro | OpenAI | 74.1% | 2026-11-11 |
| 11 | GPT-5.6 Sol Pro | OpenAI | 71.7% | 2026-11-11 |
| 12 | GPT-5.6 Sol Pro Extra High | OpenAI | 71.7% | 2026-11-11 |
| 13 | Qwen3.7 Max | Alibaba | 70.4% | 2026-11-11 |
| 14 | Grok 4.5 | xAI | 70.0% | 2026-11-11 |
| 15 | GPT-5.5 | OpenAI | 69.0% | 2026-11-11 |
| 16 | Claude Opus 4.6 | Anthropic | 67.6% | 2026-11-11 |
| 17 | Claude Opus 4.8 | Anthropic | 64.8% | 2026-11-11 |
| 18 | GPT-5.6 Sol | OpenAI | 64.8% | 2026-11-11 |
| 19 | GPT-5.6 Sol Extra High | OpenAI | 64.8% | 2026-11-11 |
| 20 | Qwen3.6 Max Preview | Alibaba | 63.0% | 2026-11-11 |
| 21 | Gemini 2.5 Pro Preview 06-05 | 62.4% | 2026-11-11 | |
| 22 | Claude Opus 4.5 | Anthropic | 62.0% | 2026-11-11 |
| 23 | Claude Opus 4.7 | Anthropic | 61.7% | 2026-11-11 |
| 24 | GPT-5 Pro | OpenAI | 61.6% | 2026-11-11 |
| 25 | GPT-5 Pro High | OpenAI | 61.6% | 2026-11-11 |
| 26 | DeepSeek V4 Flash 0731 | DeepSeek | 61.1% | 2026-11-11 |
| 27 | Gemini 3 Flash Preview | 61.1% | 2026-11-11 | |
| 28 | Kimi K3 Max | Moonshot AI | 60.7% | 2026-11-11 |
| 29 | Claude Sonnet 5 | Anthropic | 60.6% | 2026-11-11 |
| 30 | Grok 4 | xAI | 60.5% | 2026-11-11 |
| 31 | Claude Opus 4.1 | Anthropic | 60.0% | 2026-11-11 |
| 32 | Claude Opus 4 | Anthropic | 58.8% | 2026-11-11 |
| 33 | GLM-5.2 | Zhipu AI | 58.8% | 2026-11-11 |
| 34 | Claude Opus 4 Thinking | Anthropic | 58.8% | 2026-11-11 |
| 35 | Kimi K2.7 Code | Moonshot AI | 57.9% | 2026-11-11 |
| 36 | GPT-5.2 Pro | OpenAI | 57.4% | 2026-11-11 |
| 37 | GPT-5 High | OpenAI | 56.7% | 2026-11-11 |
| 38 | Grok 4.1 Fast (Reasoning) | xAI | 56.0% | 2026-11-11 |
| 39 | Grok 4.1 Fast | xAI | 56.0% | 2026-11-11 |
| 40 | GLM-5.1 | Zhipu AI | 55.1% | 2026-11-11 |
| 41 | Claude Sonnet 4.5 | Anthropic | 54.3% | 2026-11-11 |
| 42 | Claude Sonnet 4.5 Thinking | Anthropic | 54.3% | 2026-11-11 |
| 43 | GLM-5 | Zhipu AI | 53.2% | 2026-11-11 |
| 44 | GPT-5.1 High | OpenAI | 53.2% | 2026-11-11 |
| 45 | o3 High | OpenAI | 53.1% | 2026-11-11 |
| 46 | DeepSeek-V3.2-Speciale | DeepSeek | 52.6% | 2026-11-11 |
| 47 | Inkling | Thinking Machines | 50.0% | 2026-11-11 |
| 48 | GPT-5.6 Terra | OpenAI | 48.9% | 2026-11-11 |
| 49 | GPT-5.6 Terra Extra High | OpenAI | 48.9% | 2026-11-11 |
| 50 | GLM-4.7 | Zhipu AI | 47.7% | 2026-11-11 |
| 51 | Kimi K2.5 | Moonshot AI | 46.8% | 2026-11-11 |
| 52 | GPT-5.6 Luna | OpenAI | 46.8% | 2026-11-11 |
| 53 | GPT-5.6 Luna Extra High | OpenAI | 46.8% | 2026-11-11 |
| 54 | Claude Sonnet 3.7 | Anthropic | 46.4% | 2026-11-11 |
| 55 | Claude Sonnet 3.7 Thinking | Anthropic | 46.4% | 2026-11-11 |
| 56 | DeepSeek V4 Flash | DeepSeek | 46.3% | 2026-11-11 |
| 57 | MiniMax-M3 | MiniMax | 45.8% | 2026-11-11 |
| 58 | GPT-5.2 | OpenAI | 45.8% | 2026-11-11 |
| 59 | GPT-5.2 High | OpenAI | 45.8% | 2026-11-11 |
| 60 | Claude Sonnet 4 | Anthropic | 45.5% | 2026-11-11 |
| 61 | Claude Sonnet 4 Thinking | Anthropic | 45.5% | 2026-11-11 |
| 62 | o1-preview | OpenAI | 41.7% | 2026-11-11 |
| 63 | Claude Sonnet 3.5 v2 | Anthropic | 41.4% | 2026-11-11 |
| 64 | Gemini 2.5 Flash | 41.2% | 2026-11-11 | |
| 65 | DeepSeek R1 0528 | DeepSeek | 40.8% | 2026-11-11 |
| 66 | o1 High | OpenAI | 40.1% | 2026-11-11 |
| 67 | DeepSeek-V3.1 | DeepSeek | 40.0% | 2026-11-11 |
| 68 | o4-mini High | OpenAI | 38.7% | 2026-11-11 |
| 69 | o1 Medium | OpenAI | 36.7% | 2026-11-11 |
| 70 | Grok 3 | xAI | 36.1% | 2026-11-11 |
| 71 | Qwen3.6 Flash | Alibaba | 35.2% | 2026-11-11 |
| 72 | GPT-4.5 | OpenAI | 34.5% | 2026-11-11 |
| 73 | Gemini 2.0 Flash (Dec 2024) | 31.1% | 2026-11-11 | |
| 74 | Qwen3 235B-A22B | Alibaba | 31.0% | 2026-11-11 |
| 75 | DeepSeek-R1 | DeepSeek | 30.9% | 2026-11-11 |
| 76 | Gemini 2.0 Flash Thinking (Jan 2025) | 30.7% | 2026-11-11 | |
| 77 | Llama 4 Maverick | Meta | 27.7% | 2026-11-11 |
| 78 | Claude 3.5 Sonnet | Anthropic | 27.5% | 2026-11-11 |
| 79 | DeepSeek V3 0324 | DeepSeek | 27.2% | 2026-11-11 |
| 80 | Gemini 1.5 Pro (Sept 2024) | 27.1% | 2026-11-11 | |
| 81 | GPT-4.1 | OpenAI | 27.0% | 2026-11-11 |
| 82 | Kimi K2 Instruct | Moonshot AI | 26.3% | 2026-11-11 |
| 83 | GPT-4 Turbo | OpenAI | 25.1% | 2026-11-11 |
| 84 | Claude 3 Opus | Anthropic | 23.5% | 2026-11-11 |
| 85 | Llama 3.1-405B | Meta | 23.0% | 2026-11-11 |
| 86 | o3-mini High | OpenAI | 22.8% | 2026-11-11 |
| 87 | Grok-2 (Dec 2024) | xAI | 22.7% | 2026-11-11 |
| 88 | Mistral Large 2407 | Mistral | 22.5% | 2026-11-11 |
| 89 | GPT OSS 120B | OpenAI | 22.1% | 2026-11-11 |
| 90 | Llama-3.3-70B-Instruct | Meta | 19.9% | 2026-11-11 |
| 91 | DeepSeek-V3 | DeepSeek | 18.9% | 2026-11-11 |
| 92 | o1-mini Medium | OpenAI | 18.1% | 2026-11-11 |
| 93 | GPT-4o (2024-08-06) | OpenAI | 17.8% | 2026-11-11 |
| 94 | Command R+ | Cohere | 17.4% | 2026-11-11 |
| 95 | GPT-4o mini | OpenAI | 10.7% | 2026-11-11 |