# Phi 4

> Source: Benchize (https://benchize.com) — https://benchize.com/models/phi-4. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/phi-4.

Microsoft

Canonical URL: https://benchize.com/models/phi-4
Last updated: Thu, 22 Oct 2026 04:30:12 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Microsoft

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [SVGBench Arena](https://benchize.com/benchmarks/svgbench%3Aarena.md) | 1,297 | — | 108 / 191 | — | — | [svgbench](https://svgbench.ai/api/leaderboard?shape=2) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1297 \| rating \| reported \| Higher is better \| ### Configuration Reasoning effort unknown sourceModelId model_microsoft_phi_4 ### Reported statistics coverageCount 27 draws 1 judged 24 losses 17 rank 163 ratingDeviation 28 wins 6 ### Source [Original source](https://svgbench.ai/api/leaderboard?shape=2) Source svgbench Original model label Phi 4 Observed 2026-10-22T04:30:04.538Z Evidence SHA-256 `5c0a9c869593820d57d2c05f6d416e1b44a740b4835e66c66e9ad8b8edc59e2c` Source locator `/rows/157` |  |  |  |  |  |  |  |
| [Arena text style control / german](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Agerman.md) | 1,256.1 | — | 124 / 129 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1256.0970085107776 \| rating \| reported \| Higher is better \| ### Configuration category german config text_style_control ### Reported statistics lower 1230.2369088095281 rank 251 upper 1281.957108212027 variance 174.08614481456917 votes 523 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3387/row` |  |  |  |  |  |  |  |
| [Arena text / french](https://benchize.com/benchmarks/lmarena%3Atext%3Afrench.md) | 1,224.39 | — | 129 / 132 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1224.3907968900974 \| rating \| reported \| Higher is better \| ### Configuration category french config text ### Reported statistics lower 1184.7510555827944 rank 251 upper 1264.0305381974001 variance 409.039681082918 votes 205 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3088/row` |  |  |  |  |  |  |  |
| [Berkeley Function Calling Leaderboard](https://benchize.com/benchmarks/bfcl%3Aoverall.md) | 28.79% | — | 20 / 26 | $8.72 | 9.49s | [Berkeley Function Calling Leaderboard](https://api.github.com/repos/HuanzhiMao/BFCL-Result/contents/2025-12-16/score/data_overall.csv?ref=cadd76a8dc62984e7bf15ecaf358f070bae745cc) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 28.79 \| percent \| reported \| Higher is better \| \| `cost_total` \| 8.72 \| USD \| evaluation_total \| Lower is better \| \| `latency_mean` \| 9.49 \| seconds \| per_call_mean \| Lower is better \| \| `latency_sd` \| 26.73 \| seconds \| per_call_sd \| Lower is better \| \| `non_live_ast_acc` \| 69.56 \| percent \| reported_task_score \| Higher is better \| \| `non_live_simple_ast` \| 74.25 \| percent \| reported_task_score \| Higher is better \| \| `non_live_multiple_ast` \| 89.5 \| percent \| reported_task_score \| Higher is better \| \| `non_live_parallel_ast` \| 65 \| percent \| reported_task_score \| Higher is better \| \| `non_live_parallel_multiple_ast` \| 49.5 \| percent \| reported_task_score \| Higher is better \| \| `live_acc` \| 60.7 \| percent \| reported_task_score \| Higher is better \| \| `live_simple_ast` \| 65.5 \| percent \| reported_task_score \| Higher is better \| \| `live_multiple_ast` \| 59.64 \| percent \| reported_task_score \| Higher is better \| \| `live_parallel_ast` \| 81.25 \| percent \| reported_task_score \| Higher is better \| \| `live_parallel_multiple_ast` \| 41.67 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_acc` \| 3.88 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_base` \| 9 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_miss_func` \| 0 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_miss_param` \| 3.5 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_long_context` \| 3 \| percent \| reported_task_score \| Higher is better \| \| `web_search_acc` \| 4.5 \| percent \| reported_task_score \| Higher is better \| \| `web_search_base` \| 4 \| percent \| reported_task_score \| Higher is better \| \| `web_search_no_snippet` \| 5 \| percent \| reported_task_score \| Higher is better \| \| `memory_acc` \| 24.73 \| percent \| reported_task_score \| Higher is better \| \| `memory_kv` \| 17.42 \| percent \| reported_task_score \| Higher is better \| \| `memory_vector` \| 25.16 \| percent \| reported_task_score \| Higher is better \| \| `memory_recursive_summarization` \| 31.61 \| percent \| reported_task_score \| Higher is better \| \| `relevance_detection` \| 50 \| percent \| reported_task_score \| Higher is better \| \| `irrelevance_detection` \| 87.55 \| percent \| reported_task_score \| Higher is better \| \| `latency_p95` \| 23.02 \| seconds \| per_call_p95 \| Lower is better \| ### Configuration mode Prompt sourceRelease 2025-12-16 ### Reported statistics rank 70 ### Source [Original source](https://api.github.com/repos/HuanzhiMao/BFCL-Result/contents/2025-12-16/score/data_overall.csv?ref=cadd76a8dc62984e7bf15ecaf358f070bae745cc) Source Berkeley Function Calling Leaderboard Original model label Phi-4 Observed 2026-10-22T02:40:09.731Z Evidence SHA-256 `ede5081a24bc5ac7f9b3b09a52af6bcfcd35aa0f1709239fed979db77a8159a6` Source locator `/rows/70` |  |  |  |  |  |  |  |
| [Arena text style control / french](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Afrench.md) | 1,266.87 | — | 129 / 132 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1266.8652552169945 \| rating \| reported \| Higher is better \| ### Configuration category french config text_style_control ### Reported statistics lower 1226.5554951790823 rank 256 upper 1307.1750152549066 variance 422.9842958515591 votes 205 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3093/row` |  |  |  |  |  |  |  |
| [Arena text / industry writing and literature and language](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_writing_and_literature_and_language.md) | 1,197.91 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1197.9118075515876 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text ### Reported statistics lower 1190.1807755545283 rank 318 upper 1205.642839548647 variance 15.55889533881714 votes 6470 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7435/row` |  |  |  |  |  |  |  |
| [Arena text / longer query](https://benchize.com/benchmarks/lmarena%3Atext%3Alonger_query.md) | 1,217.1 | — | 146 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1217.1013358830867 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text ### Reported statistics lower 1206.3431656144946 rank 309 upper 1227.8595061516787 variance 30.128717482153267 votes 2896 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8826/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) | 1,264.57 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1264.574653310107 \| rating \| reported \| Higher is better \| ### Configuration category industry_life_and_physical_and_social_science config text_style_control ### Reported statistics lower 1254.8614831503583 rank 322 upper 1274.2878234698555 variance 24.55985576208378 votes 4245 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5819/row` |  |  |  |  |  |  |  |
| [Arena text style control / russian](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Arussian.md) | 1,246.61 | — | 146 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1246.6085356640865 \| rating \| reported \| Higher is better \| ### Configuration category russian config text_style_control ### Reported statistics lower 1235.3190176575708 rank 314 upper 1257.8980536706022 variance 33.17833738913077 votes 2576 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10697/row` |  |  |  |  |  |  |  |
| [Arena text style control / exclude ties](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexclude_ties.md) | 1,156.86 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1156.8634908830177 \| rating \| reported \| Higher is better \| ### Configuration category exclude_ties config text_style_control ### Reported statistics lower 1149.0521640558125 rank 321 upper 1164.674817710223 variance 15.883764384695429 votes 15789 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2374/row` |  |  |  |  |  |  |  |
| [Arena text style control / chinese](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Achinese.md) | 1,267.55 | — | 140 / 146 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1267.5529281956196 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text_style_control ### Reported statistics lower 1251.9102524552109 rank 310 upper 1283.1956039360282 variance 63.69801571251935 votes 1503 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/726/row` |  |  |  |  |  |  |  |
| [Arena text / german](https://benchize.com/benchmarks/lmarena%3Atext%3Agerman.md) | 1,222.46 | — | 122 / 129 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1222.4564888944456 \| rating \| reported \| Higher is better \| ### Configuration category german config text ### Reported statistics lower 1197.089950499515 rank 245 upper 1247.8230272893757 variance 167.50440397151064 votes 523 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3381/row` |  |  |  |  |  |  |  |
| [Arena text / russian](https://benchize.com/benchmarks/lmarena%3Atext%3Arussian.md) | 1,208.59 | — | 146 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1208.589978928364 \| rating \| reported \| Higher is better \| ### Configuration category russian config text ### Reported statistics lower 1197.349140646031 rank 305 upper 1219.8308172106968 variance 32.89282826848778 votes 2576 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10688/row` |  |  |  |  |  |  |  |
| [Arena text / instruction following](https://benchize.com/benchmarks/lmarena%3Atext%3Ainstruction_following.md) | 1,201.14 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1201.1428568162275 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text ### Reported statistics lower 1194.5260520815582 rank 307 upper 1207.759661550897 variance 11.397259983860016 votes 9162 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7840/row` |  |  |  |  |  |  |  |
| [Arena text style control / math](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Amath.md) | 1,264.77 | — | 140 / 149 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1264.7688498198158 \| rating \| reported \| Higher is better \| ### Configuration category math config text_style_control ### Reported statistics lower 1254.2526686355325 rank 288 upper 1275.2850310040992 variance 28.788559727601825 votes 2764 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9200/row` |  |  |  |  |  |  |  |
| [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,202.75 | — | 142 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1202.7479424854037 \| rating \| reported \| Higher is better \| ### Configuration category expert config text ### Reported statistics lower 1185.5300167141038 rank 289 upper 1219.9658682567033 variance 77.17301725817268 votes 1124 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2759/row` |  |  |  |  |  |  |  |
| [Arena text / korean](https://benchize.com/benchmarks/lmarena%3Atext%3Akorean.md) | 1,150.81 | — | 120 / 124 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1150.8122890993966 \| rating \| reported \| Higher is better \| ### Configuration category korean config text ### Reported statistics lower 1117.7344858083782 rank 239 upper 1183.890092390415 variance 284.82436533358486 votes 319 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8472/row` |  |  |  |  |  |  |  |
| [Arena text / overall](https://benchize.com/benchmarks/lmarena%3Atext.md) | 1,216.71 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1216.7128262443107 \| rating \| reported \| Higher is better \| ### Configuration category overall config text ### Reported statistics lower 1212.1308725767024 rank 315 upper 1221.294779911919 variance 5.465189239830537 votes 24126 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/314/row` |  |  |  |  |  |  |  |
| [Arena text / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_legal_and_government.md) | 1,240.93 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1240.93455870382 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text ### Reported statistics lower 1224.3648531458236 rank 299 upper 1257.5042642618164 variance 71.47158282672524 votes 1341 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5406/row` |  |  |  |  |  |  |  |
| [Arena text / non english](https://benchize.com/benchmarks/lmarena%3Atext%3Anon_english.md) | 1,196.71 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1196.7145310516053 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text ### Reported statistics lower 1190.32631187599 rank 311 upper 1203.1027502272202 variance 10.623397552995947 votes 10025 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10041/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,273.07 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1273.0701637845284 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control ### Reported statistics lower 1253.6551556040313 rank 304 upper 1292.4851719650255 variance 98.12484273374542 votes 1037 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6616/row` |  |  |  |  |  |  |  |
| [Arena text style control / non english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Anon_english.md) | 1,233.12 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1233.124427041478 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text_style_control ### Reported statistics lower 1226.6691063765263 rank 320 upper 1239.5797477064295 variance 10.847744784576571 votes 10025 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10050/row` |  |  |  |  |  |  |  |
| [Arena text style control / instruction following](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ainstruction_following.md) | 1,245.4 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1245.3950545654789 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text_style_control ### Reported statistics lower 1238.7409532358618 rank 316 upper 1252.049155895096 variance 11.526106766078094 votes 9162 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7849/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_legal_and_government.md) | 1,298.43 | — | 142 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1298.4277178018456 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text_style_control ### Reported statistics lower 1281.7748528320967 rank 298 upper 1315.0805827715944 variance 72.19078080617741 votes 1341 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5405/row` |  |  |  |  |  |  |  |
| [Arena text style control / expert](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexpert.md) | 1,268.52 | — | 141 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1268.5168458484966 \| rating \| reported \| Higher is better \| ### Configuration category expert config text_style_control ### Reported statistics lower 1251.231198396621 rank 291 upper 1285.802493300372 variance 77.7812861668369 votes 1124 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2761/row` |  |  |  |  |  |  |  |
| [Arena text / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts_english.md) | 1,229.92 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1229.9212612645829 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text ### Reported statistics lower 1220.5385050563918 rank 301 upper 1239.304017472774 variance 22.917365035411212 votes 3804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/4168/row` |  |  |  |  |  |  |  |
| [Arena text style control / overall](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aoverall.md) | 1,256.35 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1256.347619213395 \| rating \| reported \| Higher is better \| ### Configuration category overall config text_style_control ### Reported statistics lower 1251.7363987820936 rank 321 upper 1260.9588396446966 variance 5.535228895727155 votes 24126 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/320/row` |  |  |  |  |  |  |  |
| [Arena text / multi turn](https://benchize.com/benchmarks/lmarena%3Atext%3Amulti_turn.md) | 1,205.91 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1205.9113344489554 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text ### Reported statistics lower 1195.687504050148 rank 311 upper 1216.135164847763 variance 27.210159708205 votes 3517 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9626/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry software and it services](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_software_and_it_services.md) | 1,289.58 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1289.5815910822678 \| rating \| reported \| Higher is better \| ### Configuration category industry_software_and_it_services config text_style_control ### Reported statistics lower 1281.4565490507523 rank 311 upper 1297.7066331137835 variance 17.185218193219356 votes 5580 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7011/row` |  |  |  |  |  |  |  |
| [Arena text style control / japanese](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ajapanese.md) | 1,193.82 | — | 116 / 120 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1193.8157206419005 \| rating \| reported \| Higher is better \| ### Configuration category japanese config text_style_control ### Reported statistics lower 1166.838484118885 rank 237 upper 1220.792957164916 variance 189.45180057591418 votes 500 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8187/row` |  |  |  |  |  |  |  |
| [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,201.67 | — | 144 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1201.6698231314244 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text ### Reported statistics lower 1182.2163185120307 rank 301 upper 1221.1233277508181 variance 98.51435604049787 votes 1037 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6613/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_mathematical.md) | 1,278.36 | — | 138 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1278.3613750180548 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text_style_control ### Reported statistics lower 1266.6935491234276 rank 282 upper 1290.029200912682 variance 35.43918273285931 votes 2350 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6194/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) | 1,256.24 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1256.2407810021302 \| rating \| reported \| Higher is better \| ### Configuration category industry_business_and_management_and_financial_operations config text_style_control ### Reported statistics lower 1244.1589069777215 rank 312 upper 1268.322655026539 variance 37.99902244306943 votes 2438 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4594/row` |  |  |  |  |  |  |  |
| [Arena text / industry business and management and financial operations](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_business_and_management_and_financial_operations.md) | 1,198.74 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1198.7361559773244 \| rating \| reported \| Higher is better \| ### Configuration category industry_business_and_management_and_financial_operations config text ### Reported statistics lower 1186.6825841015016 rank 307 upper 1210.7897278531475 variance 37.821203284271746 votes 2438 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/4589/row` |  |  |  |  |  |  |  |
| [Arena text style control / spanish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aspanish.md) | 1,270.12 | — | 132 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1270.1157357242796 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text_style_control ### Reported statistics lower 1218.4164529171862 rank 256 upper 1321.8150185313727 variance 695.7814641586737 votes 127 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/11020/row` |  |  |  |  |  |  |  |
| [Arena text style control / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts_english.md) | 1,289.28 | — | 144 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1289.2799212758432 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text_style_control ### Reported statistics lower 1279.893986070886 rank 304 upper 1298.6658564808004 variance 22.9328970538682 votes 3804 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4171/row` |  |  |  |  |  |  |  |
| [Arena text / industry life and physical and social science](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_life_and_physical_and_social_science.md) | 1,206.48 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1206.4772711517014 \| rating \| reported \| Higher is better \| ### Configuration category industry_life_and_physical_and_social_science config text ### Reported statistics lower 1196.8345748441511 rank 317 upper 1216.1199674592517 variance 24.204760852504315 votes 4245 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5814/row` |  |  |  |  |  |  |  |
| [Arena text / english](https://benchize.com/benchmarks/lmarena%3Atext%3Aenglish.md) | 1,230.98 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1230.9802162052526 \| rating \| reported \| Higher is better \| ### Configuration category english config text ### Reported statistics lower 1225.2208590240084 rank 312 upper 1236.7395733864969 variance 8.634791283576726 votes 14101 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1948/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) | 1,229.38 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1229.3800530556691 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text_style_control ### Reported statistics lower 1221.587407346321 rank 320 upper 1237.1726987650172 variance 15.807881845378907 votes 6470 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7437/row` |  |  |  |  |  |  |  |
| [Arena text style control / korean](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Akorean.md) | 1,195.83 | — | 121 / 124 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1195.8311263933113 \| rating \| reported \| Higher is better \| ### Configuration category korean config text_style_control ### Reported statistics lower 1161.9451594330524 rank 246 upper 1229.7170933535701 variance 298.9121608296393 votes 319 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8479/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) | 1,202.35 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1202.3494628613344 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text_style_control ### Reported statistics lower 1193.272550845055 rank 323 upper 1211.4263748776136 variance 21.44766756510166 votes 4486 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5015/row` |  |  |  |  |  |  |  |
| [Arena text / chinese](https://benchize.com/benchmarks/lmarena%3Atext%3Achinese.md) | 1,211.58 | — | 140 / 146 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1211.5781520001751 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text ### Reported statistics lower 1196.0435646801964 rank 303 upper 1227.1127393201537 variance 62.82077056300112 votes 1503 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/719/row` |  |  |  |  |  |  |  |
| [Arena text / japanese](https://benchize.com/benchmarks/lmarena%3Atext%3Ajapanese.md) | 1,158.35 | — | 115 / 120 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1158.3525106097422 \| rating \| reported \| Higher is better \| ### Configuration category japanese config text ### Reported statistics lower 1131.7433567100852 rank 229 upper 1184.9616645093993 variance 184.31723579629525 votes 500 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8179/row` |  |  |  |  |  |  |  |
| [BuzzBench](https://benchize.com/benchmarks/eq-bench%3Abuzzbench.md) | 38.67% | CI 37.7–39.64% | 12 / 15 | — | — | [eq bench · buzzbench](https://eqbench.com/buzzbench.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 38.67 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol buzzbench sourceModelVersion microsoft/phi-4 ### Reported statistics Uncertainty Confidence interval: 37.7–39.64% confidenceIntervalHigh 39.64 confidenceIntervalLow 37.7 ### Source [Original source](https://eqbench.com/buzzbench.js) Source eq bench Original model label microsoft/phi-4 Source version buzzbench Observed 2026-10-21T16:35:02.926Z Evidence SHA-256 `4d0f19f1f87e08772b741a28b1ac39111587b832035dd1a783af3de67d51be1c` Source locator `/leaderboardDataBuzzbench/19` |  |  |  |  |  |  |  |
| [Arena text / math](https://benchize.com/benchmarks/lmarena%3Atext%3Amath.md) | 1,245.81 | — | 141 / 149 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1245.8124779112807 \| rating \| reported \| Higher is better \| ### Configuration category math config text ### Reported statistics lower 1235.3113197877758 rank 283 upper 1256.3136360347858 variance 28.706365753760274 votes 2764 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9195/row` |  |  |  |  |  |  |  |
| [Arena text / creative writing](https://benchize.com/benchmarks/lmarena%3Atext%3Acreative_writing.md) | 1,181.4 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1181.3981519660397 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text ### Reported statistics lower 1171.7490845931418 rank 317 upper 1191.0472193389376 variance 24.23675627216523 votes 4062 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1538/row` |  |  |  |  |  |  |  |
| [Arena text style control / longer query](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Alonger_query.md) | 1,267.07 | — | 146 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1267.071195303239 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text_style_control ### Reported statistics lower 1256.2146933860963 rank 315 upper 1277.9276972203813 variance 30.6819985267006 votes 2896 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8832/row` |  |  |  |  |  |  |  |
| [Arena text / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts.md) | 1,219.55 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1219.5524932101657 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text ### Reported statistics lower 1211.790596261837 rank 301 upper 1227.314390158495 variance 15.683376302753677 votes 5747 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3751/row` |  |  |  |  |  |  |  |
| [Arena text style control / multi turn](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Amulti_turn.md) | 1,241.85 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1241.8466415069383 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text_style_control ### Reported statistics lower 1231.5886718455813 rank 315 upper 1252.1046111682954 variance 27.39218262772056 votes 3517 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9630/row` |  |  |  |  |  |  |  |
| [Arena text / industry entertainment and sports and media](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_entertainment_and_sports_and_media.md) | 1,161.87 | — | 145 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1161.873305569136 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text ### Reported statistics lower 1152.8714565299867 rank 320 upper 1170.8751546082854 variance 21.094404471318246 votes 4486 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5012/row` |  |  |  |  |  |  |  |
| [Arena text style control / creative writing](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acreative_writing.md) | 1,210 | — | 144 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1209.9990512478107 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text_style_control ### Reported statistics lower 1200.2578403000289 rank 320 upper 1219.740262195593 variance 24.70186331765538 votes 4062 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1541/row` |  |  |  |  |  |  |  |
| [Arena text style control / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts.md) | 1,277.97 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1277.9684427258742 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text_style_control ### Reported statistics lower 1270.1786026273037 rank 310 upper 1285.7582828244447 variance 15.796501171482335 votes 5747 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3760/row` |  |  |  |  |  |  |  |
| [Arena text / industry software and it services](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_software_and_it_services.md) | 1,227.86 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1227.8574560527659 \| rating \| reported \| Higher is better \| ### Configuration category industry_software_and_it_services config text ### Reported statistics lower 1219.8045632490735 rank 306 upper 1235.9103488564579 variance 16.881368650475277 votes 5580 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7006/row` |  |  |  |  |  |  |  |
| [Arena text / spanish](https://benchize.com/benchmarks/lmarena%3Atext%3Aspanish.md) | 1,234.22 | — | 131 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1234.2153481377964 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text ### Reported statistics lower 1182.8464418004908 rank 252 upper 1285.584254475102 variance 686.9173044562451 votes 127 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/11016/row` |  |  |  |  |  |  |  |
| [Arena text / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_mathematical.md) | 1,253.98 | — | 139 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1253.9758971606311 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text ### Reported statistics lower 1242.3495598690567 rank 281 upper 1265.6022344522053 variance 35.187600629551 votes 2350 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6193/row` |  |  |  |  |  |  |  |
| [Arena text / coding](https://benchize.com/benchmarks/lmarena%3Atext%3Acoding.md) | 1,231.18 | — | 144 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1231.179345945576 \| rating \| reported \| Higher is better \| ### Configuration category coding config text ### Reported statistics lower 1221.3380345166975 rank 300 upper 1241.0206573744545 variance 25.212143396781922 votes 3305 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1109/row` |  |  |  |  |  |  |  |
| [Arena text style control / english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aenglish.md) | 1,271.92 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1271.9169220473937 \| rating \| reported \| Higher is better \| ### Configuration category english config text_style_control ### Reported statistics lower 1266.1330859578673 rank 319 upper 1277.7007581369198 variance 8.708347914676589 votes 14101 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1955/row` |  |  |  |  |  |  |  |
| [Arena text style control / coding](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acoding.md) | 1,305.78 | — | 144 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1305.7808053990195 \| rating \| reported \| Higher is better \| ### Configuration category coding config text_style_control ### Reported statistics lower 1295.8592684712896 rank 308 upper 1315.7023423267492 variance 25.624873154443538 votes 3305 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1117/row` |  |  |  |  |  |  |  |
| [Arena text / exclude ties](https://benchize.com/benchmarks/lmarena%3Atext%3Aexclude_ties.md) | 1,100.17 | — | 146 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1100.167380684833 \| rating \| reported \| Higher is better \| ### Configuration category exclude_ties config text ### Reported statistics lower 1092.481547776025 rank 315 upper 1107.8532135936407 variance 15.377498564838964 votes 15789 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Phi-4 Observed 2026-10-22T00:45:04.817Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2368/row` |  |  |  |  |  |  |  |

---
Cite as: Phi 4 — Benchize, https://benchize.com/models/phi-4 (retrieved 2026-10-22)