# Creative Writing v2 > Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v2. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v2. LLM-judged creative writing quality on the legacy v2 rubric. Original published scores are retained without the page’s optional vocabulary-complexity or repetitive-language slider penalties. Canonical URL: https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v2 Last updated: Wed, 21 Oct 2026 18:52:20 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. LLM-judged creative writing quality on the legacy v2 rubric. Original published scores are retained without the page’s optional vocabulary-complexity or repetitive-language slider penalties. Writing & creativity · Models tested: 18 · Score publishers: [eq bench](https://eqbench.com/creative_writing_v2.html) ## Model results | # | Model | Score | ± | | | --- | --- | --- | --- | --- | | 1 | [DeepSeek R1](https://benchize.com/models/deepseek-r1.md) | 87.11% | CI 86.78–87.45% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 87.11 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion deepseek-ai/DeepSeek-R1 ### Reported statistics Uncertainty Confidence interval: 86.78–87.45% confidenceIntervalHigh 87.45 confidenceIntervalLow 86.78 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label deepseek-ai/DeepSeek-R1 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/4` | | | | | | 2 | [Gemma 3 27B](https://benchize.com/models/gemma-3-27b-it.md) | 85.25% | CI 85.25–85.25% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 85.25 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion google/gemma-3-27b-it ### Reported statistics Uncertainty Confidence interval: 85.25–85.25% confidenceIntervalHigh 85.25 confidenceIntervalLow 85.25 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label google/gemma-3-27b-it Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/82` | | | | | | 3 | [GPT-4o (2024-11-20)](https://benchize.com/models/gpt-4o-2024-11-20.md) | 84.59% | CI 84.17–85.01% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 84.59 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gpt-4o-2024-11-20 ### Reported statistics Uncertainty Confidence interval: 84.17–85.01% confidenceIntervalHigh 85.01 confidenceIntervalLow 84.17 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gpt-4o-2024-11-20 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/12` | | | | | | 4 | [Gemini 1.5 Pro 002](https://benchize.com/models/gemini-1-5-pro-002.md) | 81.44% | CI 81.13–81.76% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 81.44 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gemini-1.5-pro-002 ### Reported statistics Uncertainty Confidence interval: 81.13–81.76% confidenceIntervalHigh 81.76 confidenceIntervalLow 81.13 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gemini-1.5-pro-002 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/18` | | | | | | 5 | [Gemini 2.0 Flash 001](https://benchize.com/models/gemini-2-0-flash-001.md) | 81.41% | CI 81.18–81.65% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 81.41 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gemini-2.0-flash-001 ### Reported statistics Uncertainty Confidence interval: 81.18–81.65% confidenceIntervalHigh 81.65 confidenceIntervalLow 81.18 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gemini-2.0-flash-001 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/75` | | | | | | 6 | [DeepSeek V3](https://benchize.com/models/deepseek-chat.md) | 81.19% | CI 80.76–81.61% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 81.19 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion deepseek-ai/DeepSeek-V3 ### Reported statistics Uncertainty Confidence interval: 80.76–81.61% confidenceIntervalHigh 81.61 confidenceIntervalLow 80.76 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label deepseek-ai/DeepSeek-V3 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/11` | | | | | | 7 | [Gemini 1.5 Pro 001](https://benchize.com/models/gemini-1-5-pro-001.md) | 80.27% | CI 79.76–80.77% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 80.27 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gemini-1.5-pro-001 ### Reported statistics Uncertainty Confidence interval: 79.76–80.77% confidenceIntervalHigh 80.77 confidenceIntervalLow 79.76 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gemini-1.5-pro-001 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/68` | | | | | | 8 | [GPT-4o Mini](https://benchize.com/models/gpt-4o-mini.md) | 78.41% | CI 77.62–79.21% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 78.41 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gpt-4o-mini ### Reported statistics Uncertainty Confidence interval: 77.62–79.21% confidenceIntervalHigh 79.21 confidenceIntervalLow 77.62 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gpt-4o-mini Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/41` | | | | | | 9 | [Mistral Nemo](https://benchize.com/models/mistral-nemo.md) | 77.46% | CI 76.82–78.1% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 77.46 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion mistralai/Mistral-Nemo-Instruct-2407 ### Reported statistics Uncertainty Confidence interval: 76.82–78.1% confidenceIntervalHigh 78.1 confidenceIntervalLow 76.82 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label mistralai/Mistral-Nemo-Instruct-2407 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/39` | | | | | | 10 | [Gemma 2 27B](https://benchize.com/models/gemma-2-27b-it.md) | 77.22% | CI 76.4–78.04% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 77.22 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion google/gemma-2-27b-it ### Reported statistics Uncertainty Confidence interval: 76.4–78.04% confidenceIntervalHigh 78.04 confidenceIntervalLow 76.4 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label google/gemma-2-27b-it Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/46` | | | | | | 11 | [GPT-4o](https://benchize.com/models/gpt-4o.md) | 75.61% | CI 74.87–76.34% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 75.61 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion gpt-4o ### Reported statistics Uncertainty Confidence interval: 74.87–76.34% confidenceIntervalHigh 76.34 confidenceIntervalLow 74.87 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label gpt-4o Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/69` | | | | | | 12 | [DeepSeek R1 Distill Qwen](https://benchize.com/models/deepseek-r1-distill-qwen.md) | 73.05% | — | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 73.05 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion deepseek-ai/deepseek-r1-distill-qwen-32b ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label deepseek-ai/deepseek-r1-distill-qwen-32b Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/7` | | | | | | 13 | [Qwen2.5 72B Instruct](https://benchize.com/models/qwen-2-5-72b-instruct.md) | 72.16% | CI 70.06–74.26% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 72.16 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion Qwen/Qwen2.5-72B-Instruct ### Reported statistics Uncertainty Confidence interval: 70.06–74.26% confidenceIntervalHigh 74.26 confidenceIntervalLow 70.06 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label Qwen/Qwen2.5-72B-Instruct Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/19` | | | | | | 14 | [Mistral Small 3](https://benchize.com/models/mistral-small-24b-instruct-2501.md) | 67.55% | — | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 67.55 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion mistralai/mistral-small-24b-instruct-2501 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label mistralai/mistral-small-24b-instruct-2501 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/9` | | | | | | 15 | [Mistral Small 3.1 24B](https://benchize.com/models/mistral-small-3-1-24b-instruct.md) | 66.25% | CI 66.25–66.25% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 66.25 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion mistralai/Mistral-Small-3.1-24B-Instruct-2503 ### Reported statistics Uncertainty Confidence interval: 66.25–66.25% confidenceIntervalHigh 66.25 confidenceIntervalLow 66.25 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label mistralai/Mistral-Small-3.1-24B-Instruct-2503 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/83` | | | | | | 16 | [Mixtral 8x22B Instruct](https://benchize.com/models/mixtral-8x22b-instruct.md) | 63.64% | CI 62.49–64.8% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 63.64 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion mistralai/Mixtral-8x22B-Instruct-v0.1 ### Reported statistics Uncertainty Confidence interval: 62.49–64.8% confidenceIntervalHigh 64.8 confidenceIntervalLow 62.49 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label mistralai/Mixtral-8x22B-Instruct-v0.1 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/55` | | | | | | 17 | [Llama 3.2 3B Instruct](https://benchize.com/models/llama-3-2-3b-instruct.md) | 62.5% | CI 61.13–63.88% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 62.5 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion meta-llama/Llama-3.2-3B-Instruct ### Reported statistics Uncertainty Confidence interval: 61.13–63.88% confidenceIntervalHigh 63.88 confidenceIntervalLow 61.13 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label meta-llama/Llama-3.2-3B-Instruct Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/3` | | | | | | 18 | [Skyfall 36B V2](https://benchize.com/models/skyfall-36b-v2.md) | 62.06% | CI 62.06–62.06% | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 62.06 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v2 sourceModelVersion thedrummer/skyfall-36b-v2 ### Reported statistics Uncertainty Confidence interval: 62.06–62.06% confidenceIntervalHigh 62.06 confidenceIntervalLow 62.06 ### Source [Original source](https://eqbench.com/creative_writing_v2.js) Source eq bench Original model label thedrummer/skyfall-36b-v2 Source version creative-writing-v2 Observed 2026-10-21T16:35:01.751Z Evidence SHA-256 `264c33d486559c25267ca29861754404dadb674022dcdf6530388cd9951af31a` Source locator `/leaderboardDataCreativeWriting/79` | | | | | --- Cite as: Creative Writing v2 — Benchize, https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v2 (retrieved 2026-10-21)