# Short Story Writing — Pairwise Comparison > Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise. Models write short stories to the same constrained creative briefs. Independent evaluators make pairwise comparisons of matched stories in both orders. The current pooled-panel Thurstone rating measures relative writing quality; negative ratings are valid. Story/evaluator bootstrap intervals describe this observed comparison pool, rather than worst-case uncertainty about missing judgments. Canonical URL: https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise Last updated: Wed, 21 Oct 2026 18:05:10 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. Models write short stories to the same constrained creative briefs. Independent evaluators make pairwise comparisons of matched stories in both orders. The current pooled-panel Thurstone rating measures relative writing quality; negative ratings are valid. Story/evaluator bootstrap intervals describe this observed comparison pool, rather than worst-case uncertainty about missing judgments. Writing & creativity · Models tested: 50 · Score publishers: [lechmazur short story](https://github.com/lechmazur/writing) ## Model results | # | Model | Score | ± | | | --- | --- | --- | --- | --- | | 1 | [Claude Fable 5.1](https://benchize.com/models/claude-fable-5-1-high.md)High | 3.76 | 95% CI 3.67–3.85 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.756462887159027 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.9340839066310491 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 3.67–3.85 bootstrapSamples 300 confidenceInterval95 ["3.6748593284624396","3.8502845844526776"] sourceRank 1 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Fable 5.1 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/0` | | | | | | 2 | [Claude Opus 5](https://benchize.com/models/claude-opus-5-xhigh.md)Extra high | 3.73 | 95% CI 3.67–3.78 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.7303542555060156 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.9324651908129759 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort xhigh thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 3.67–3.78 bootstrapSamples 300 confidenceInterval95 ["3.6719256557595905","3.7840126992302263"] sourceRank 2 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 5 (xhigh) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/1` | | | | | | 3 | [Claude Opus 5.5](https://benchize.com/models/claude-opus-5-5-high.md)High | 3.71 | 95% CI 3.62–3.81 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.714673136037459 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.9314822187137942 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 3.62–3.81 bootstrapSamples 300 confidenceInterval95 ["3.6214890390860743","3.808658567911917"] sourceRank 3 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 5.5 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/2` | | | | | | 4 | [GPT-6 Astra](https://benchize.com/models/gpt-6-astra-high.md)High | 3.44 | 95% CI 3.37–3.51 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.4432356224193557 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.913183161554768 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 3.37–3.51 bootstrapSamples 300 confidenceInterval95 ["3.3742839351451726","3.507684861495963"] sourceRank 4 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-6 Astra (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/3` | | | | | | 5 | [Claude Opus 5](https://benchize.com/models/claude-opus-5-high.md)High | 3.37 | 95% CI 3.27–3.47 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.3722357746297056 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.9079955615224807 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 3.27–3.47 bootstrapSamples 300 confidenceInterval95 ["3.2710203620630973","3.4706627597654123"] sourceRank 5 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 5 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/4` | | | | | | 6 | [GLM 5.3](https://benchize.com/models/glm-5-3-max.md)Max | 3.03 | 95% CI 2.94–3.11 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.0277077443758276 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8804767850780137 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort max thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.94–3.11 bootstrapSamples 300 confidenceInterval95 ["2.9420163968638193","3.107084573676175"] sourceRank 6 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GLM-5.3 (max) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/5` | | | | | | 7 | [Claude Fable 5](https://benchize.com/models/claude-fable-5-high.md)High | 2.85 | 95% CI 2.79–2.92 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 2.8529094861340387 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8650486749750985 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.79–2.92 bootstrapSamples 300 confidenceInterval95 ["2.7903520559217703","2.919944096045654"] sourceRank 7 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Fable 5 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/6` | | | | | | 8 | [GPT-5.6 Sol](https://benchize.com/models/gpt-5-6-sol-xhigh.md)Extra high | 2.47 | 95% CI 2.39–2.54 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 2.4689133916762596 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8278026604335277 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort xhigh thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.39–2.54 bootstrapSamples 300 confidenceInterval95 ["2.394621536001434","2.543300550622248"] sourceRank 8 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.6 Sol (xhigh) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/7` | | | | | | 9 | [GPT-5.5](https://benchize.com/models/gpt-5-5-xhigh.md)Extra high | 2.43 | 95% CI 2.37–2.5 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 2.432634818748467 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8240518700529466 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort xhigh thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.37–2.5 bootstrapSamples 300 confidenceInterval95 ["2.3656110696462105","2.496184424290147"] sourceRank 9 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.5 (xhigh) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/8` | | | | | | 10 | [Kimi K3](https://benchize.com/models/kimi-k3.md) | 2.42 | 95% CI 2.34–2.5 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 2.41992507841589 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8227285609736585 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: 2.34–2.5 bootstrapSamples 300 confidenceInterval95 ["2.342189338415047","2.5014128352949645"] sourceRank 10 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Kimi K3 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/9` | | | | | | 11 | [GPT-5.6 Sol](https://benchize.com/models/gpt-5-6-sol-high.md)High | 2.28 | 95% CI 2.2–2.37 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 2.280340865497532 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8078821627327935 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.2–2.37 bootstrapSamples 300 confidenceInterval95 ["2.2032537397212977","2.3714404005146013"] sourceRank 11 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.6 Sol (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/10` | | | | | | 12 | [GPT-5.4](https://benchize.com/models/gpt-5-4-medium.md)Medium | 1.97 | 95% CI 1.85–2.11 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.9683840352890751 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.7726805363876615 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort medium thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.85–2.11 bootstrapSamples 300 confidenceInterval95 ["1.846033944979465","2.1084425627830083"] sourceRank 12 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.4 (medium) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/11` | | | | | | 13 | [GPT-5.4](https://benchize.com/models/gpt-5-4-xhigh.md)Extra high | 1.96 | 95% CI 1.85–2.08 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.9617969313963528 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.7719079539717636 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort xhigh thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.85–2.08 bootstrapSamples 300 confidenceInterval95 ["1.8480658401593317","2.0754132132070033"] sourceRank 13 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.4 (xhigh) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/12` | | | | | | 14 | [Muse Wonder](https://benchize.com/models/muse-wonder-high.md)High | 1.96 | 95% CI 1.87–2.05 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.959756656511701 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.7716684154149746 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.87–2.05 bootstrapSamples 300 confidenceInterval95 ["1.8695921467725831","2.0527870600884155"] sourceRank 14 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Muse Wonder (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/13` | | | | | | 15 | [Claude Opus 4.7](https://benchize.com/models/claude-opus-4-7-high.md)High | 1.76 | 95% CI 1.66–1.87 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.7644170415584985 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.7482150489255891 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.66–1.87 bootstrapSamples 300 confidenceInterval95 ["1.6578978968868754","1.8708247570658847"] sourceRank 15 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 4.7 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/14` | | | | | | 16 | [Claude Sonnet 4.6](https://benchize.com/models/claude-sonnet-4-6-thinking-budget-16384.md)Thinking · 16384 tokens | 1.66 | 95% CI 1.55–1.75 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.6551236833288474 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.734654373626279 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 reasoningBudget 16384 thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.55–1.75 bootstrapSamples 300 confidenceInterval95 ["1.5486216127951014","1.754013039176927"] sourceRank 16 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Sonnet 4.6 Thinking 16K Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/15` | | | | | | 17 | [MiMo-V2.6-Pro](https://benchize.com/models/mimo-v2-6-pro-thinking.md)Thinking | 1.6 | 95% CI 1.45–1.71 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.5960626325851228 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.7271992198936259 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 1.45–1.71 bootstrapSamples 300 confidenceInterval95 ["1.453328634777985","1.712998710360439"] sourceRank 17 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Xiaomi MiMo V2.6 Pro (thinking) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/16` | | | | | | 18 | [Claude Opus 4.6](https://benchize.com/models/claude-opus-4-6-thinking-budget-16384.md)Thinking · 16384 tokens | 1.13 | 95% CI 0.97–1.29 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 1.1279392312470806 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.6651502440805583 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 reasoningBudget 16384 thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.97–1.29 bootstrapSamples 300 confidenceInterval95 ["0.9683533210425165","1.2889868478225666"] sourceRank 18 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 4.6 Thinking 16K Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/17` | | | | | | 19 | [Gemini 4 Argon](https://benchize.com/models/gemini-4-argon-high.md)High | 0.8 | 95% CI 0.68–0.92 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.7997952223558088 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.6188677077131939 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.68–0.92 bootstrapSamples 300 confidenceInterval95 ["0.6832085942797734","0.9176309890990524"] sourceRank 19 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 4 Argon (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/18` | | | | | | 20 | [Claude Opus 4.8](https://benchize.com/models/claude-opus-4-8-xhigh.md)Extra high | 0.76 | 95% CI 0.65–0.86 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.7594311754943414 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.6130392888797267 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort xhigh thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.65–0.86 bootstrapSamples 300 confidenceInterval95 ["0.6493780646751726","0.8616691120467034"] sourceRank 20 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 4.8 (xhigh) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/19` | | | | | | 21 | [Muse Spark 1.1](https://benchize.com/models/muse-spark-1-1-high.md)High | 0.76 | 95% CI 0.66–0.86 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.7553636837747042 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.6124504636527286 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.66–0.86 bootstrapSamples 300 confidenceInterval95 ["0.6563973526983181","0.8625153663175532"] sourceRank 21 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Muse Spark 1.1 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/20` | | | | | | 22 | [Muse Spark 1.3](https://benchize.com/models/muse-spark-1-3-high.md)High | 0.56 | 95% CI 0.48–0.66 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.5591149600614204 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.583739122391948 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.48–0.66 bootstrapSamples 300 confidenceInterval95 ["0.4758272563503359","0.6580261863090415"] sourceRank 22 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Muse Spark 1.3 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/21` | | | | | | 23 | [DeepSeek V4 Pro](https://benchize.com/models/deepseek-v4-pro-high.md)High | 0.52 | 95% CI 0.4–0.63 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.5167277285142252 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5774666113422205 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.4–0.63 bootstrapSamples 300 confidenceInterval95 ["0.39850336831052324","0.630659697863716"] sourceRank 23 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label DeepSeek V4 Pro (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/22` | | | | | | 24 | [GLM-5.2](https://benchize.com/models/glm-5-2-max.md)Max | 0.46 | 95% CI 0.35–0.55 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.4574569700670838 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5686581363155423 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort max thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.35–0.55 bootstrapSamples 300 confidenceInterval95 ["0.3452674156005926","0.5479126902897905"] sourceRank 24 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GLM-5.2 (max) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/23` | | | | | | 25 | [Grok 4.7](https://benchize.com/models/grok-4-7-high.md)High | 0.43 | 95% CI 0.22–0.62 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.4319651721712296 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5648570274979469 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.22–0.62 bootstrapSamples 300 confidenceInterval95 ["0.21813989983659846","0.6225823456459567"] sourceRank 25 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Grok 4.7 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/24` | | | | | | 26 | [GPT-5.2](https://benchize.com/models/gpt-5-2-medium.md)Medium | 0.33 | 95% CI 0.18–0.47 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.33259015486618937 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5499731475522392 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort medium thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.18–0.47 bootstrapSamples 300 confidenceInterval95 ["0.18370605578987498","0.47450018520360404"] sourceRank 26 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-5.2 (medium) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/25` | | | | | | 27 | [Claude Opus 4.8](https://benchize.com/models/claude-opus-4-8-high.md)High | 0.25 | 95% CI 0.13–0.36 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.24531646783522268 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.536825753931911 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.13–0.36 bootstrapSamples 300 confidenceInterval95 ["0.13239058345541016","0.356124723013479"] sourceRank 27 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Claude Opus 4.8 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/26` | | | | | | 28 | [Qwen3.8 Max](https://benchize.com/models/qwen3-8-max.md) | 0.18 | 95% CI 0.07–0.3 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.17620397982599334 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5263726183470008 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: 0.07–0.3 bootstrapSamples 300 confidenceInterval95 ["0.06699429561637041","0.2965953890718493"] sourceRank 28 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Qwen 3.8 Max Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/27` | | | | | | 29 | [Gemini 3.8 Flash](https://benchize.com/models/gemini-3-8-flash-high.md)High | 0.15 | 95% CI 0.00824–0.28 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.1482517060222879 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5221360708134836 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 0.00824–0.28 bootstrapSamples 300 confidenceInterval95 ["0.008239475449616123","0.27563638384343925"] sourceRank 29 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 3.8 Flash (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/28` | | | | | | 30 | [Kimi K2.6](https://benchize.com/models/kimi-k2-6.md) | 0.08 | 95% CI -0.03–0.19 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 0.07844723446558423 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.5115378923409997 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.03–0.19 bootstrapSamples 300 confidenceInterval95 ["-0.026638259912426777","0.1929822243811043"] sourceRank 30 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Kimi K2.6 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/29` | | | | | | 31 | [Muse Spark 1.2](https://benchize.com/models/muse-spark-1-2-high.md)High | -0.05 | 95% CI -0.16–0.05 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.04840932181096462 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.49222980866925503 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -0.16–0.05 bootstrapSamples 300 confidenceInterval95 ["-0.1622524315788904","0.04994169841090842"] sourceRank 31 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Muse Spark 1.2 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/30` | | | | | | 32 | [MiniMax M3](https://benchize.com/models/minimax-m3.md) | -0.09 | 95% CI -0.22–0.03 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.09146209144780505 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.4856691024565312 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.22–0.03 bootstrapSamples 300 confidenceInterval95 ["-0.2241683518996273","0.0344453191239487"] sourceRank 32 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label MiniMax-M3 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/31` | | | | | | 33 | [Mistral Medium 3.1](https://benchize.com/models/mistral-medium-3-1.md) | -0.39 | 95% CI -0.54–-0.26 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.3861839020278999 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.4408162006877146 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.54–-0.26 bootstrapSamples 300 confidenceInterval95 ["-0.53883571599254","-0.2610992956716161"] sourceRank 33 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Mistral Medium 3.1 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/32` | | | | | | 34 | [DeepSeek V4.1 Flash](https://benchize.com/models/deepseek-v4-1-flash-high.md)High | -0.49 | 95% CI -0.61–-0.35 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.48788265529811664 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.4254277762172336 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -0.61–-0.35 bootstrapSamples 300 confidenceInterval95 ["-0.6105971728903827","-0.35266875687308785"] sourceRank 34 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label DeepSeek V4.1 Flash (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/33` | | | | | | 35 | [DeepSeek V4 Pro](https://benchize.com/models/deepseek-v4-pro.md) | -0.54 | 95% CI -0.65–-0.42 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.5411854791095495 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.417394212854644 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.65–-0.42 bootstrapSamples 300 confidenceInterval95 ["-0.6511833537397125","-0.4168640401380341"] sourceRank 35 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label DeepSeek V4 Pro Preview Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/34` | | | | | | 36 | [Qwen3.8 27B](https://benchize.com/models/qwen3-8-27b.md) | -0.64 | 95% CI -0.79–-0.48 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.6365328760986696 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.403091590819266 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.79–-0.48 bootstrapSamples 300 confidenceInterval95 ["-0.7929545796461278","-0.4766647149259913"] sourceRank 36 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Qwen3.8-27B Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/35` | | | | | | 37 | [MiMo-V2.5-Pro](https://benchize.com/models/mimo-v2-5-pro.md) | -0.68 | 95% CI -0.81–-0.55 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.6768701594333522 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.3970706892506185 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -0.81–-0.55 bootstrapSamples 300 confidenceInterval95 ["-0.8123617489744717","-0.5537118790796263"] sourceRank 37 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Xiaomi MiMo V2.5 Pro Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/36` | | | | | | 38 | [Gemini 3.7 Flash](https://benchize.com/models/gemini-3-7-flash-high.md)High | -0.71 | 95% CI -0.83–-0.6 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.7060157386399251 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.3927325750567854 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -0.83–-0.6 bootstrapSamples 300 confidenceInterval95 ["-0.8319882408203993","-0.6015755844436229"] sourceRank 39 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 3.7 Flash (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/38` | | | | | | 39 | [Qwen3.6 Max Preview](https://benchize.com/models/qwen3-6-max-preview.md) | -0.94 | 95% CI -1.06–-0.8 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.9405244734076871 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.35826855312938216 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -1.06–-0.8 bootstrapSamples 300 confidenceInterval95 ["-1.059633172955714","-0.8011492881452411"] sourceRank 40 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Qwen 3.6 Max Preview Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/39` | | | | | | 40 | [GLM-5.1](https://benchize.com/models/glm-5-1.md) | -1.04 | 95% CI -1.23–-0.85 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -1.0428779767813396 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.34351562140767145 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -1.23–-0.85 bootstrapSamples 300 confidenceInterval95 ["-1.2314791244968841","-0.8536609707894092"] sourceRank 41 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GLM-5.1 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/40` | | | | | | 41 | [MiMo-V2-Pro](https://benchize.com/models/mimo-v2-pro.md) | -1.23 | 95% CI -1.45–-0.98 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -1.2265817150557672 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.31756733639259227 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -1.45–-0.98 bootstrapSamples 300 confidenceInterval95 ["-1.4501729016360971","-0.981165702831619"] sourceRank 43 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Xiaomi MiMo V2 Pro Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/42` | | | | | | 42 | [Gemini 3.5 Flash](https://benchize.com/models/gemini-3-5-flash.md) | -1.99 | 95% CI -2.08–-1.9 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -1.9883734732594616 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.21932130001084554 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -2.08–-1.9 bootstrapSamples 300 confidenceInterval95 ["-2.076116928787029","-1.9000101605550352"] sourceRank 49 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 3.5 Flash Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/48` | | | | | | 43 | [Gemini 3.1 Pro Preview](https://benchize.com/models/gemini-3-1-pro-preview.md) | -2.17 | 95% CI -2.27–-2.07 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -2.169608572429032 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.19859238875107632 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -2.27–-2.07 bootstrapSamples 300 confidenceInterval95 ["-2.2710552684989196","-2.071592808223834"] sourceRank 50 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 3.1 Pro Preview Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/49` | | | | | | 44 | [Mistral Medium 3.5](https://benchize.com/models/mistral-medium-3-5.md) | -2.55 | 95% CI -2.66–-2.41 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -2.552648050997538 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.15854122258598707 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -2.66–-2.41 bootstrapSamples 300 confidenceInterval95 ["-2.6611589719735584","-2.4082123190646105"] sourceRank 52 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Mistral Medium 3.5 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/51` | | | | | | 45 | [DeepSeek V3.2](https://benchize.com/models/deepseek-v3-2.md) | -2.88 | 95% CI -3.14–-2.63 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -2.8769381221964996 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.12874783306808807 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -3.14–-2.63 bootstrapSamples 300 confidenceInterval95 ["-3.1423481035149323","-2.6315448071315264"] sourceRank 54 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label DeepSeek V3.2 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/53` | | | | | | 46 | [Grok 4.6](https://benchize.com/models/grok-4-6-high.md)High | -2.88 | 95% CI -3–-2.74 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -2.879594243062891 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.12851949910241295 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -3–-2.74 bootstrapSamples 300 confidenceInterval95 ["-3.0045967872712365","-2.738949813585302"] sourceRank 55 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Grok 4.6 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/54` | | | | | | 47 | [gpt-oss-120b](https://benchize.com/models/gpt-oss-120b.md) | -3.16 | 95% CI -3.27–-3.05 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -3.159886408755964 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.10584418878561416 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -3.27–-3.05 bootstrapSamples 300 confidenceInterval95 ["-3.270748435601263","-3.0486142137822956"] sourceRank 56 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GPT-OSS-120B Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/55` | | | | | | 48 | [MiniMax M2.7](https://benchize.com/models/minimax-m2-7.md) | -3.77 | 95% CI -3.9–-3.62 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -3.767908021633996 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.065966154898169 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -3.9–-3.62 bootstrapSamples 300 confidenceInterval95 ["-3.9040077549166776","-3.6196598038203303"] sourceRank 57 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label MiniMax-M2.7 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/56` | | | | | | 49 | [Grok 4.3](https://benchize.com/models/grok-4-3.md) | -4.26 | 95% CI -4.43–-4.09 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -4.257115065216755 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.04226035615277023 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 ### Reported statistics Uncertainty 95% confidence interval: -4.43–-4.09 bootstrapSamples 300 confidenceInterval95 ["-4.432702198393286","-4.085431485757633"] sourceRank 58 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Grok 4.3 Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/57` | | | | | | 50 | [Grok 4.5](https://benchize.com/models/grok-4-5-high.md)High | -5.09 | 95% CI -5.19–-4.99 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -5.088657194243192 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.015743435556097907 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -5.19–-4.99 bootstrapSamples 300 confidenceInterval95 ["-5.186858848362259","-4.990649443075917"] sourceRank 59 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Grok 4.5 (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/58` | | | | | --- Cite as: Short Story Writing — Pairwise Comparison — Benchize, https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise (retrieved 2026-10-21)