# DeepSeek V4 ProHigh · 20260813 tokens > Source: Benchize (https://benchize.com) — https://benchize.com/models/deepseek-v4-pro-high-budget-20260813. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/deepseek-v4-pro-high-budget-20260813. DeepSeek Canonical URL: https://benchize.com/models/deepseek-v4-pro-high-budget-20260813 Last updated: Thu, 22 Oct 2026 04:40:22 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. DeepSeek ## Benchmarks | Benchmark | Score | Rank | | | --- | --- | --- | --- | | [Arena text style control / industry software and it services](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_software_and_it_services.md) | 1,492.79 | 43 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1492.7938720119012 \| rating \| reported \| Higher is better \| ### Configuration category industry_software_and_it_services config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1483.305889352942 rank 66 upper 1502.2818546708604 variance 23.434278262143373 votes 4090 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6766/row` | | | | | [Arena text style control / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts.md) | 1,482.66 | 41 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1482.663790199744 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1474.8906827485287 rank 61 upper 1490.4368976509593 variance 15.728711999370551 votes 6526 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3511/row` | | | | | [Arena text / multi turn](https://benchize.com/benchmarks/lmarena%3Atext%3Amulti_turn.md) | 1,453.85 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1453.846006970733 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1438.5241897912802 rank 58 upper 1469.167824150186 variance 61.111700694519655 votes 1569 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9373/row` | | | | | [Arena text style control / industry writing and literature and language](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_writing_and_literature_and_language.md) | 1,453.69 | 37 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1453.693857803972 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1441.7537921560897 rank 48 upper 1465.6339234518541 variance 37.11224676097845 votes 2636 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7165/row` | | | | | [Arena text / industry life and physical and social science](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_life_and_physical_and_social_science.md) | 1,456.54 | 46 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1456.5435699953698 \| rating \| reported \| Higher is better \| ### Configuration category industry_life_and_physical_and_social_science config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1441.45364337763 rank 66 upper 1471.6334966131096 variance 59.275888655140186 votes 1527 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5563/row` | | | | | [Arena text style control / non english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Anon_english.md) | 1,447.76 | 41 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1447.755366509572 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1439.8272049891923 rank 60 upper 1455.6835280299517 variance 16.36246749662453 votes 6040 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9790/row` | | | | | [Arena text / industry software and it services](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_software_and_it_services.md) | 1,461.88 | 52 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1461.8797496034304 \| rating \| reported \| Higher is better \| ### Configuration category industry_software_and_it_services config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1452.4612872391876 rank 77 upper 1471.2982119676735 variance 23.092121365140592 votes 4090 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6777/row` | | | | | [Arena text / chinese](https://benchize.com/benchmarks/lmarena%3Atext%3Achinese.md) | 1,481 | 49 / 146 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1481.000792219238 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1458.6608725999665 rank 74 upper 1503.3407118385096 variance 129.91731315899713 votes 711 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/490/row` | | | | | [Arena text / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts.md) | 1,457.51 | 44 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1457.5086408254406 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1449.8025942769896 rank 61 upper 1465.214687373892 variance 15.458490167068316 votes 6526 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3511/row` | | | | | [Arena text / russian](https://benchize.com/benchmarks/lmarena%3Atext%3Arussian.md) | 1,447.9 | 42 / 150 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1447.896521597393 \| rating \| reported \| Higher is better \| ### Configuration category russian config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1431.2591192176017 rank 58 upper 1464.5339239771843 variance 72.05678125610333 votes 1233 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10441/row` | | | | | [Arena text / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts_english.md) | 1,462.89 | 42 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1462.8934336961684 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1450.8708705856843 rank 57 upper 1474.9159968066524 variance 37.62685752790387 votes 2490 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3924/row` | | | | | [Arena text / math](https://benchize.com/benchmarks/lmarena%3Atext%3Amath.md) | 1,440.61 | 55 / 149 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1440.6078877501768 \| rating \| reported \| Higher is better \| ### Configuration category math config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1412.786247233009 rank 82 upper 1468.429528267345 variance 201.497326197202 votes 407 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8994/row` | | | | | [Arena text / industry entertainment and sports and media](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_entertainment_and_sports_and_media.md) | 1,423.5 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1423.5027341824525 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1411.101690952425 rank 57 upper 1435.90377741248 variance 40.03319581731729 votes 2463 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/4749/row` | | | | | [Arena text style control / instruction following](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ainstruction_following.md) | 1,461.47 | 38 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1461.4701453825917 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1451.1901596719235 rank 50 upper 1471.7501310932596 variance 27.50988911874966 votes 3482 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7583/row` | | | | | [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,458.88 | 54 / 148 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1458.8776397519441 \| rating \| reported \| Higher is better \| ### Configuration category expert config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1440.4810592710935 rank 74 upper 1477.2742202327947 variance 88.10042985889923 votes 1028 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2544/row` | | | | | [Arena text style control / spanish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aspanish.md) | 1,454.83 | 40 / 134 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1454.8277138281105 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1422.4498955458903 rank 65 upper 1487.2055321103303 variance 272.8971376887047 votes 299 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10829/row` | | | | | [Arena text / english](https://benchize.com/benchmarks/lmarena%3Atext%3Aenglish.md) | 1,457.36 | 38 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1457.3609332605831 \| rating \| reported \| Higher is better \| ### Configuration category english config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1447.6329108688635 rank 52 upper 1467.0889556523027 variance 24.635021243491487 votes 3765 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1688/row` | | | | | [Arena text style control / creative writing](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acreative_writing.md) | 1,446.15 | 33 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1446.1469800396153 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1432.2943204734174 rank 47 upper 1459.9996396058136 variance 49.953985195213974 votes 1998 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1268/row` | | | | | [Arena text style control / chinese](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Achinese.md) | 1,495.44 | 45 / 146 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1495.4439090390347 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1473.2915000563296 rank 68 upper 1517.59631802174 variance 127.74553799547608 votes 711 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/484/row` | | | | | [Arena text style control / expert](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexpert.md) | 1,485.42 | 49 / 148 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1485.4209459839835 \| rating \| reported \| Higher is better \| ### Configuration category expert config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1467.0719180103988 rank 66 upper 1503.7699739575683 variance 87.64556469059268 votes 1028 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2536/row` | | | | | [Arena text / instruction following](https://benchize.com/benchmarks/lmarena%3Atext%3Ainstruction_following.md) | 1,448.29 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1448.294325416126 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1438.0404069078525 rank 53 upper 1458.5482439243995 variance 27.370551054173802 votes 3482 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7586/row` | | | | | [Arena text style control / industry entertainment and sports and media](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_entertainment_and_sports_and_media.md) | 1,434.97 | 39 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1434.9675057995235 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1422.4609752225756 rank 60 upper 1447.474036376471 variance 40.71716354982716 votes 2463 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4752/row` | | | | | [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,445.51 | 49 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1445.5105175947147 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1422.4521838417054 rank 84 upper 1468.568851347724 variance 138.40751138628463 votes 623 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6396/row` | | | | | [Arena text style control / multi turn](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Amulti_turn.md) | 1,470.92 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1470.9224787070357 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1455.5633210943374 rank 61 upper 1486.281636319734 variance 61.40993137838281 votes 1569 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9376/row` | | | | | [Arena text style control / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts_english.md) | 1,489.07 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1489.0672477506228 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1476.9727172033226 rank 56 upper 1501.1617782979229 variance 38.078676874412835 votes 2490 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3923/row` | | | | | [Arena text style control / industry life and physical and social science](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_life_and_physical_and_social_science.md) | 1,485.95 | 38 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1485.9479128976159 \| rating \| reported \| Higher is better \| ### Configuration category industry_life_and_physical_and_social_science config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1470.718128464238 rank 50 upper 1501.1776973309936 variance 60.37975277455844 votes 1527 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5547/row` | | | | | [Arena text style control / exclude ties](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexclude_ties.md) | 1,469.62 | 39 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1469.6210228955078 \| rating \| reported \| Higher is better \| ### Configuration category exclude_ties config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1460.8003870481184 rank 57 upper 1478.4416587428975 variance 20.25366413746838 votes 7172 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2110/row` | | | | | [Arena text / korean](https://benchize.com/benchmarks/lmarena%3Atext%3Akorean.md) | 1,448.57 | 18 / 124 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1448.5678117421505 \| rating \| reported \| Higher is better \| ### Configuration category korean config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1407.7384130947703 rank 21 upper 1489.397210389531 variance 433.9600843633347 votes 197 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8254/row` | | | | | [Arena text / overall](https://benchize.com/benchmarks/lmarena%3Atext.md) | 1,445.56 | 41 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1445.558129330793 \| rating \| reported \| Higher is better \| ### Configuration category overall config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1438.9921141566563 rank 60 upper 1452.12414450493 variance 11.222964316250078 votes 9805 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/59/row` | | | | | [Arena webdev / overall](https://benchize.com/benchmarks/lmarena%3Awebdev%3Aoverall.md) | 1,582.38 | 26 / 90 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1582.3778052448145 \| rating \| reported \| Higher is better \| ### Configuration category overall config webdev reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1572.9059336648802 rank 36 upper 1591.8496768247487 variance 23.354760629857694 votes 4714 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/35/row` | | | | | [Arena text / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_mathematical.md) | 1,442.69 | 57 / 148 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1442.6943410973029 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1416.6277115971345 rank 86 upper 1468.7609705974712 variance 176.87790113451388 votes 513 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5998/row` | | | | | [Arena text style control / english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aenglish.md) | 1,478.5 | 34 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1478.4960386360674 \| rating \| reported \| Higher is better \| ### Configuration category english config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1468.7075828108104 rank 44 upper 1488.2844944613241 variance 24.942052463728512 votes 3765 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1680/row` | | | | | [Arena text style control / overall](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aoverall.md) | 1,464.57 | 39 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1464.568498397921 \| rating \| reported \| Higher is better \| ### Configuration category overall config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1457.9594496768666 rank 57 upper 1471.1775471189758 variance 11.370556613016387 votes 9805 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/56/row` | | | | | [Arena text style control / longer query](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Alonger_query.md) | 1,474.93 | 39 / 150 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1474.9292016892437 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1465.6742215308363 rank 55 upper 1484.1841818476507 variance 22.297429630401577 votes 4568 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8572/row` | | | | | [Arena text style control / industry business and management and financial operations](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_business_and_management_and_financial_operations.md) | 1,467.09 | 36 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1467.0863942903713 \| rating \| reported \| Higher is better \| ### Configuration category industry_business_and_management_and_financial_operations config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1453.0243614863698 rank 46 upper 1481.148427094373 variance 51.475435716133084 votes 1804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4328/row` | | | | | [Arena text style control / french](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Afrench.md) | 1,472.86 | 48 / 132 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1472.8619680057095 \| rating \| reported \| Higher is better \| ### Configuration category french config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1439.6027868928877 rank 76 upper 1506.1211491185318 variance 287.95652379155877 votes 315 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2913/row` | | | | | [Arena text / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_legal_and_government.md) | 1,458.88 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1458.8803739369193 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1438.6220174339765 rank 53 upper 1479.1387304398622 variance 106.83467592818694 votes 825 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5160/row` | | | | | [Arena text style control / coding](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acoding.md) | 1,505.01 | 45 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1505.0108861028507 \| rating \| reported \| Higher is better \| ### Configuration category coding config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1493.6180543993526 rank 68 upper 1516.403717806349 variance 33.78836537958236 votes 2804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/877/row` | | | | | [Arena text style control / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_legal_and_government.md) | 1,480.36 | 33 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1480.3624861124842 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1459.8645691394947 rank 40 upper 1500.8604030854735 variance 109.37631244883273 votes 825 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5147/row` | | | | | [Arena text / industry business and management and financial operations](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_business_and_management_and_financial_operations.md) | 1,442.43 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1442.4338132891733 \| rating \| reported \| Higher is better \| ### Configuration category industry_business_and_management_and_financial_operations config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1428.4852016301331 rank 52 upper 1456.3824249482132 variance 50.64840632068861 votes 1804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/4334/row` | | | | | [Arena text style control / math](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Amath.md) | 1,457.15 | 49 / 149 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1457.154629666686 \| rating \| reported \| Higher is better \| ### Configuration category math config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1429.3555911827643 rank 69 upper 1484.9536681506079 variance 201.17007020029158 votes 407 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8981/row` | | | | | [Arena text / non english](https://benchize.com/benchmarks/lmarena%3Atext%3Anon_english.md) | 1,430.75 | 46 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1430.745342876463 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1422.8472701049823 rank 70 upper 1438.6434156479438 variance 16.2385063631459 votes 6040 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9800/row` | | | | | [Arena text style control / russian](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Arussian.md) | 1,459.85 | 43 / 150 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1459.8542043110233 \| rating \| reported \| Higher is better \| ### Configuration category russian config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1443.049066087944 rank 60 upper 1476.6593425341025 variance 73.51703711502189 votes 1233 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10443/row` | | | | | [Arena text / coding](https://benchize.com/benchmarks/lmarena%3Atext%3Acoding.md) | 1,469.05 | 47 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1469.0521825990365 \| rating \| reported \| Higher is better \| ### Configuration category coding config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1457.7274339502274 rank 71 upper 1480.376931247846 variance 33.38573649880562 votes 2804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/880/row` | | | | | [Arena text / exclude ties](https://benchize.com/benchmarks/lmarena%3Atext%3Aexclude_ties.md) | 1,442.12 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1442.1161460873332 \| rating \| reported \| Higher is better \| ### Configuration category exclude_ties config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1433.4481864152929 rank 61 upper 1450.7841057593735 variance 19.558591770233235 votes 7172 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2114/row` | | | | | [Arena text / spanish](https://benchize.com/benchmarks/lmarena%3Atext%3Aspanish.md) | 1,435.16 | 53 / 134 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1435.1604229132645 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1402.9149428984565 rank 90 upper 1467.405902928073 variance 270.67086487667035 votes 299 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10854/row` | | | | | [Arena text / french](https://benchize.com/benchmarks/lmarena%3Atext%3Afrench.md) | 1,446.19 | 57 / 132 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1446.186482501377 \| rating \| reported \| Higher is better \| ### Configuration category french config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1413.1046886655226 rank 94 upper 1479.2682763372316 variance 284.8930925674287 votes 315 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2931/row` | | | | | [Arena webdev / webdev-html](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev-html.md) | 1,565.64 | 26 / 89 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1565.6369185101405 \| rating \| reported \| Higher is better \| ### Configuration category webdev-html config webdev reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1538.6740931018626 rank 36 upper 1592.5997439184184 variance 189.2494460908771 votes 548 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/395/row` | | | | | [Arena text / creative writing](https://benchize.com/benchmarks/lmarena%3Atext%3Acreative_writing.md) | 1,438.99 | 31 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1438.9917648496512 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1425.1831421507338 rank 45 upper 1452.8003875485686 variance 49.63688789630235 votes 1998 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1266/row` | | | | | [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,476.91 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1476.9057731245261 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1453.5249091321448 rank 65 upper 1500.2866371169077 variance 142.30656283111284 votes 623 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6377/row` | | | | | [Arena webdev / webdev](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev.md) | 1,582.38 | 26 / 90 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1582.3778052448145 \| rating \| reported \| Higher is better \| ### Configuration category webdev config webdev reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1572.9059336648802 rank 36 upper 1591.8496768247487 variance 23.354760629857694 votes 4714 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/247/row` | | | | | [Arena text / longer query](https://benchize.com/benchmarks/lmarena%3Atext%3Alonger_query.md) | 1,457.59 | 40 / 150 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1457.591608852711 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1448.3801659743704 rank 54 upper 1466.803051731052 variance 22.08813991284806 votes 4568 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8571/row` | | | | | [Arena text style control / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_mathematical.md) | 1,459.47 | 54 / 148 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1459.474993755147 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1433.5592014570725 rank 77 upper 1485.3907860532217 variance 174.83677992820046 votes 513 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5989/row` | | | | | [Arena text style control / korean](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Akorean.md) | 1,466.27 | 10 / 124 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1466.2709338271407 \| rating \| reported \| Higher is better \| ### Configuration category korean config text_style_control reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1425.6827397419615 rank 12 upper 1506.8591279123198 variance 428.8478872197059 votes 197 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8245/row` | | | | | [Arena webdev / webdev-react](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev-react.md) | 1,588.1 | 24 / 81 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1588.099402399057 \| rating \| reported \| Higher is better \| ### Configuration category webdev-react config webdev reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1576.4521286258364 rank 34 upper 1599.7466761722776 variance 35.31444502738147 votes 3099 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/540/row` | | | | | [Arena text / industry writing and literature and language](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_writing_and_literature_and_language.md) | 1,442.93 | 40 / 151 | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1442.9332161373309 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text reasoningBudget 20260813 Reasoning effort high thinkingEnabled true ### Reported statistics lower 1431.0175138471152 rank 52 upper 1454.8489184275466 variance 36.96094835227049 votes 2636 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4-Pro (High, 20260813) Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7169/row` | | | | --- Cite as: DeepSeek V4 ProHigh · 20260813 tokens — Benchize, https://benchize.com/models/deepseek-v4-pro-high-budget-20260813 (retrieved 2026-10-22)