EQ-Bench Longform Creative Writing
Planning and writing an eight-chapter novella, each chapter graded by a judge model on a rubric; overall score out of 100.
| Category | General |
|---|---|
| Metric | Rubric score (0-100) |
| Source | EQ-Bench (original) |
| License | None stated |
| Benchmark page | eqbench.com |
| Status | Scored |
| Weight | 0.6 within General |
| Fit | Discrimination 0.63, noise 0.21 |
Results
| # | Model | Provider | Result | Date |
|---|---|---|---|---|
| 1 | Claude Opus 5 | Anthropic | 86.3% | 2026-11-11 |
| 2 | Claude Fable 5.1 | Anthropic | 85.3% | 2026-11-11 |
| 3 | Claude Opus 5.5 | Anthropic | 84.6% | 2026-11-11 |
| 4 | Grok 4.7 | xAI | 83.0% | 2026-11-11 |
| 5 | Claude Fable 5 | Anthropic | 83.0% | 2026-11-11 |
| 6 | Muse Spark 1.3 | Meta | 82.8% | 2026-11-11 |
| 7 | GPT-6 Astra | OpenAI | 82.8% | 2026-11-11 |
| 8 | GPT-6 Sol | OpenAI | 82.8% | 2026-11-11 |
| 9 | Claude Opus 4.7 | Anthropic | 81.8% | 2026-11-11 |
| 10 | GLM-5.3 | Zhipu AI | 81.8% | 2026-11-11 |
| 11 | GPT-5.6 Sol | OpenAI | 81.7% | 2026-11-11 |
| 12 | Muse Spark 1.2 | Meta | 81.5% | 2026-11-11 |
| 13 | Gemini 4 Argon | 80.9% | 2026-11-11 | |
| 14 | Claude Opus 4.8 | Anthropic | 80.8% | 2026-11-11 |
| 15 | Kimi K3.1 Pro | Moonshot AI | 79.9% | 2026-11-11 |
| 16 | Claude Sonnet 4.6 | Anthropic | 79.9% | 2026-11-11 |
| 17 | Kimi K3 | Moonshot AI | 79.6% | 2026-11-11 |
| 18 | Kimi K2.6 | Moonshot AI | 78.5% | 2026-11-11 |
| 19 | Claude Sonnet 5 | Anthropic | 78.3% | 2026-11-11 |
| 20 | GPT-5.4 | OpenAI | 78.3% | 2026-11-11 |
| 21 | GPT-5.5 | OpenAI | 78.2% | 2026-11-11 |
| 22 | GPT-5.6 Terra | OpenAI | 78.0% | 2026-11-11 |
| 23 | GLM-5.2 | Zhipu AI | 77.9% | 2026-11-11 |
| 24 | Claude Opus 4.6 | Anthropic | 77.7% | 2026-11-11 |
| 25 | Gemini 3.8 Flash | 76.8% | 2026-11-11 | |
| 26 | DeepSeek V4 Pro | DeepSeek | 75.6% | 2026-11-11 |
| 27 | GPT-5.6 Luna | OpenAI | 75.5% | 2026-11-11 |
| 28 | Kimi K2.5 | Moonshot AI | 74.9% | 2026-11-11 |
| 29 | DeepSeek V4.1 Flash | DeepSeek | 74.0% | 2026-11-11 |
| 30 | MiMo-V2.5-Pro | Xiaomi | 73.7% | 2026-11-11 |
| 31 | GLM-5.1 | Zhipu AI | 73.5% | 2026-11-11 |
| 32 | Claude Opus 4.5 | Anthropic | 73.1% | 2026-11-11 |
| 33 | GPT-5 High | OpenAI | 73.1% | 2026-11-11 |
| 34 | GPT-5.4 mini | OpenAI | 73.0% | 2026-11-11 |
| 35 | GPT-5 | OpenAI | 72.8% | 2026-11-11 |
| 36 | Inkling | Thinking Machines | 72.5% | 2026-11-11 |
| 37 | Gemini 3.5 Flash | 71.8% | 2026-11-11 | |
| 38 | GPT-5.2 | OpenAI | 71.6% | 2026-11-11 |
| 39 | Claude Sonnet 4.5 | Anthropic | 71.4% | 2026-11-11 |
| 40 | GLM-5 | Zhipu AI | 70.9% | 2026-11-11 |
| 41 | MiMo-V2-Pro | Xiaomi | 70.9% | 2026-11-11 |
| 42 | Muse Spark 1.1 | Meta | 70.6% | 2026-11-11 |
| 43 | Claude Opus 4.1 | Anthropic | 70.5% | 2026-11-11 |
| 44 | Kimi K2 Thinking | Moonshot AI | 69.0% | 2026-11-11 |
| 45 | Gemini 3 Pro Preview | 68.8% | 2026-11-11 | |
| 46 | Qwen3.5 397B-A17B | Alibaba | 68.3% | 2026-11-11 |
| 47 | Gemini 3.1 Pro Preview | 68.2% | 2026-11-11 | |
| 48 | Kimi K2 Instruct | Moonshot AI | 68.1% | 2026-11-11 |
| 49 | GPT-5 Chat (latest) | OpenAI | 66.8% | 2026-11-11 |
| 50 | DeepSeek V3.2 | DeepSeek | 66.8% | 2026-11-11 |
| 51 | Qwen3 Max | Alibaba | 66.2% | 2026-11-11 |
| 52 | GLM-4.7 | Zhipu AI | 66.0% | 2026-11-11 |
| 53 | DeepSeek V4 Flash | DeepSeek | 66.0% | 2026-11-11 |
| 54 | MiniMax-M2.5 | MiniMax | 65.3% | 2026-11-11 |
| 55 | Claude Haiku 4.5 (latest) | Anthropic | 65.0% | 2026-11-11 |
| 56 | Kimi K2 Instruct 0905 | Moonshot AI | 63.8% | 2026-11-11 |
| 57 | o3 | OpenAI | 62.7% | 2026-11-11 |
| 58 | GPT-5.3 Chat (latest) | OpenAI | 62.6% | 2026-11-11 |
| 59 | DeepSeek V3.1 | DeepSeek | 62.3% | 2026-11-11 |
| 60 | DeepSeek V4 Flash 0731 | DeepSeek | 61.2% | 2026-11-11 |
| 61 | DeepSeek-V3.2-Exp | DeepSeek | 60.2% | 2026-11-11 |
| 62 | DeepSeek V3 0324 | DeepSeek | 60.1% | 2026-11-11 |
| 63 | DeepSeek-R1 | DeepSeek | 59.5% | 2026-11-11 |
| 64 | DeepSeek R1 0528 | DeepSeek | 59.1% | 2026-11-11 |
| 65 | Qwen3.5 27B | Alibaba | 59.0% | 2026-11-11 |
| 66 | Gemini 2.5 Pro Preview 06-05 | 58.8% | 2026-11-11 | |
| 67 | Claude Sonnet 4 | Anthropic | 58.4% | 2026-11-11 |
| 68 | Claude 3.5 Sonnet | Anthropic | 58.4% | 2026-11-11 |
| 69 | MiniMax-M2.1 | MiniMax | 57.8% | 2026-11-11 |
| 70 | GPT-5.4 nano | OpenAI | 57.8% | 2026-11-11 |
| 71 | Claude Sonnet 3.7 | Anthropic | 57.7% | 2026-11-11 |
| 72 | Qwen3 235B-A22B Instruct 2507 | Alibaba | 57.4% | 2026-11-11 |
| 73 | GLM-4.6 | Zhipu AI | 57.3% | 2026-11-11 |
| 74 | Gemma 4 31B IT | 56.5% | 2026-11-11 | |
| 75 | GPT-5 Mini | OpenAI | 56.2% | 2026-11-11 |
| 76 | Gemini 3 Flash Preview | 55.7% | 2026-11-11 | |
| 77 | GLM-4.5 | Zhipu AI | 55.5% | 2026-11-11 |
| 78 | Qwen3.5 122B-A10B | Alibaba | 54.3% | 2026-11-11 |
| 79 | Qwen3.8 27B | Alibaba | 53.3% | 2026-11-11 |
| 80 | MiMo-V2-Flash | Xiaomi | 53.1% | 2026-11-11 |
| 81 | Hermes 4 405B | Nousresearch | 51.2% | 2026-11-11 |
| 82 | Gemma 4 26B A4B IT | 50.7% | 2026-11-11 | |
| 83 | Gemini 3.1 Flash Lite Preview | 49.0% | 2026-11-11 | |
| 84 | DeepSeek-V3.2-Speciale | DeepSeek | 48.8% | 2026-11-11 |
| 85 | Gemma 4 12B IT | 48.7% | 2026-11-11 | |
| 86 | Gemini 2.5 Flash Preview 05-20 | 48.6% | 2026-11-11 | |
| 87 | GLM-4.7-Flash | Zhipu AI | 47.8% | 2026-11-11 |
| 88 | MiniMax-M2 | MiniMax | 46.4% | 2026-11-11 |
| 89 | Mistral Medium 3.1 | Mistral | 46.2% | 2026-11-11 |
| 90 | Grok 4 | xAI | 45.7% | 2026-11-11 |
| 91 | Qwen3.5 35B-A3B | Alibaba | 44.5% | 2026-11-11 |
| 92 | Gemma 3 27B IT | 41.9% | 2026-11-11 | |
| 93 | Gemma 3 12B IT | 41.8% | 2026-11-11 | |
| 94 | Reka Flash 3 | Reka | 41.2% | 2026-11-11 |
| 95 | Qwen3 32B | Alibaba | 40.5% | 2026-11-11 |
| 96 | Grok 4.1 Fast | xAI | 40.1% | 2026-11-11 |
| 97 | Nova 2 Lite | Amazon | 38.8% | 2026-11-11 |
| 98 | QwQ 32B | Alibaba | 36.9% | 2026-11-11 |
| 99 | Qwen3-14B | Alibaba | 35.9% | 2026-11-11 |
| 100 | GPT OSS 120B | OpenAI | 35.8% | 2026-11-11 |
| 101 | Qwen3 30B A3B | Alibaba | 35.6% | 2026-11-11 |
| 102 | Trinity Large Preview | Arcee AI | 35.0% | 2026-11-11 |
| 103 | MiniMax-M2 Her | MiniMax | 34.8% | 2026-11-11 |
| 104 | Gemma 3 4B IT | 34.4% | 2026-11-11 | |
| 105 | GPT-4.1 nano | OpenAI | 32.4% | 2026-11-11 |
| 106 | Ministral 3 14B | Mistral | 31.9% | 2026-11-11 |
| 107 | Llama 4 Maverick | Meta | 31.9% | 2026-11-11 |
| 108 | Qwen3-Next 80B-A3B Instruct | Alibaba | 31.5% | 2026-11-11 |
| 109 | Llama 4 Scout | Meta | 31.1% | 2026-11-11 |
| 110 | Mistral Nemo | Mistral | 28.7% | 2026-11-11 |
| 111 | GPT-5 Nano | OpenAI | 28.0% | 2026-11-11 |
| 112 | Solar Pro 3 | Upstage | 27.0% | 2026-11-11 |
| 113 | Phi-4-multimodal-instruct | Microsoft | 22.8% | 2026-11-11 |
| 114 | GPT OSS 20B | OpenAI | 21.2% | 2026-11-11 |
| 115 | Qwen3-4B | Alibaba | 17.0% | 2026-11-11 |