# DeepSeek V4.1 FlashMax

> Source: Benchize (https://benchize.com) — https://benchize.com/models/deepseek-v4-1-flash-max. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/deepseek-v4-1-flash-max.

DeepSeek

Canonical URL: https://benchize.com/models/deepseek-v4-1-flash-max
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

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

DeepSeek

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate arbitration international dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__arbitration_international_dispute_resolution.md) | 90.32% | SE ±1.77 pp | 10 / 49 | $0.21 | 553.88source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.323 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.210863 \| USD \| per_test \| Lower is better \| \| `latency` \| 553.876 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.77 pp sourcePopulation 81 stderr 1.773 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__arbitration_international_dispute_resolution/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text / longer query](https://benchize.com/benchmarks/lmarena%3Atext%3Alonger_query.md) | 1,473.47 | — | 29 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1473.468680562838 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1465.0466150988905 rank 39 upper 1481.8907460267856 variance 18.464648455140903 votes 5836 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8556/row` |  |  |  |  |  |  |  |
| [Chess Puzzles](https://benchize.com/benchmarks/epoch-ai%3Achess_puzzles.md) | 48% | SE ±5.02 pp | 16 / 63 | — | — | [epoch ai · Chess Puzzles](https://epoch.ai/data/benchmark_data.zip#file=chess_puzzles.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.48 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.48 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Chess Puzzles reasoningBudget Not reported Reasoning effort max sourceModelVersion deepseek-v4.1-flash_max ### Reported statistics Uncertainty Standard error: ±5.02 pp standardError 0.05021167315686782 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=chess_puzzles.csv) Source epoch ai Original model label deepseek-v4.1-flash Source version Chess Puzzles Evaluated 2026-10-13T12:39:44.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `fa0a449c370b0cceaf273d9472078cce3e5aef6be7009bddf115e6df7f930c54` Source locator `/rows/148` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / comparables](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Acomparables.md) | 55.82% | SE ±6.48 pp | 10 / 18 | $0.46 | 1,133.07source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 55.818 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.461555 \| USD \| per_test \| Lower is better \| \| `latency` \| 1133.071 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.48 pp sourcePopulation 28 stderr 6.481 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/comparables/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_legal_and_government.md) | 1,442.96 | — | 50 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1442.9559108611434 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1425.5497614912802 rank 80 upper 1460.3620602310064 variance 78.86952588267978 votes 1145 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5187/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate bankruptcy restructuring](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__bankruptcy_restructuring.md) | 80.28% | SE ±2.35 pp | 26 / 49 | $0.21 | 624.69source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.277 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.211492 \| USD \| per_test \| Lower is better \| \| `latency` \| 624.686 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.35 pp sourcePopulation 81 stderr 2.345 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__bankruptcy_restructuring/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / grommet](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Agrommet.md) | 25% | SE ±9.93 pp | 10 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 25 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.93 pp sourcePopulation 20 stderr 9.934 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/grommet/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / health](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Ahealth.md) | 73.33% | SE ±8.21 pp | 11 / 50 | $0.27 | 706.36source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 73.333 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.272236 \| USD \| per_test \| Lower is better \| \| `latency` \| 706.355 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.21 pp sourcePopulation 82 stderr 8.212 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/health/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / structured finance securitization](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Astructured_finance_securitization.md) | 0% | SE ±0 pp | 12 / 49 | $0.19 | 564.45source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.186047 \| USD \| per_test \| Lower is better \| \| `latency` \| 564.454 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/structured_finance_securitization/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / eslint](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Aeslint.md) | 36.36% | SE ±15.21 pp | 12 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 36.364 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±15.21 pp sourcePopulation 20 stderr 15.212 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/eslint/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text style control / longer query](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Alonger_query.md) | 1,480.64 | — | 35 / 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` \| 1480.6362690349747 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1472.1678911421534 rank 45 upper 1489.104646927796 variance 18.668278766728594 votes 5836 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8562/row` |  |  |  |  |  |  |  |
| [Arena text / german](https://benchize.com/benchmarks/lmarena%3Atext%3Agerman.md) | 1,469.91 | — | 24 / 129 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1469.908999386951 \| rating \| reported \| Higher is better \| ### Configuration category german config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1433.85576402605 rank 32 upper 1505.9622347478517 variance 338.37035372765746 votes 239 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3168/row` |  |  |  |  |  |  |  |
| [Arena text style control / german](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Agerman.md) | 1,494.61 | — | 11 / 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` \| 1494.608768965928 \| rating \| reported \| Higher is better \| ### Configuration category german config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1458.240761945196 rank 14 upper 1530.97677598666 variance 344.3045979134095 votes 239 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3150/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass criminal](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__criminal.md) | 79.26% | SE ±3.09 pp | 33 / 50 | $0.23 | 682.99source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.256 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.233636 \| USD \| per_test \| Lower is better \| \| `latency` \| 682.99 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.09 pp sourcePopulation 82 stderr 3.087 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__criminal/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text style control / non english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Anon_english.md) | 1,460.2 | — | 33 / 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` \| 1460.2014689110492 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1453.0409798523135 rank 43 upper 1467.361957969785 variance 13.347169904325574 votes 8493 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9773/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,505.46 | — | 15 / 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` \| 1505.4614024021398 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1485.0454435376607 rank 16 upper 1525.8773612666191 variance 108.50340868180578 votes 910 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6328/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass.md) | 84.5% | SE ±1.43 pp | 21 / 50 | $0.25 | 701.78source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.503 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.24793 \| USD \| per_test \| Lower is better \| \| `latency` \| 701.783 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.43 pp sourcePopulation 82 stderr 1.427 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / scratch gui](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Ascratch_gui.md) | 16.67% | SE ±16.67 pp | 6 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 16.667 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±16.67 pp sourcePopulation 20 stderr 16.667 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/scratch_gui/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text / korean](https://benchize.com/benchmarks/lmarena%3Atext%3Akorean.md) | 1,436.2 | — | 26 / 124 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1436.2029772627736 \| rating \| reported \| Higher is better \| ### Configuration category korean config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1397.3185695702985 rank 32 upper 1475.0873849552486 variance 393.5997318126665 votes 232 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8265/row` |  |  |  |  |  |  |  |
| [Arena text style control / coding](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acoding.md) | 1,524.03 | — | 24 / 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` \| 1524.026424940282 \| rating \| reported \| Higher is better \| ### Configuration category coding config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1513.9505614347756 rank 27 upper 1534.1022884457884 variance 26.4282477361692 votes 3680 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/836/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass constitutional civil rights](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__constitutional_civil_rights.md) | 85.71% | SE ±2.75 pp | 17 / 50 | $0.19 | 518.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 85.714 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.191689 \| USD \| per_test \| Lower is better \| \| `latency` \| 518.106 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.75 pp sourcePopulation 82 stderr 2.752 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__constitutional_civil_rights/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass rule source lookup](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__rule_source_lookup.md) | 56.89% | SE ±8.71 pp | 20 / 46 | $0.11 | 302.33source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 56.89 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.112383 \| USD \| per_test \| Lower is better \| \| `latency` \| 302.327 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.71 pp sourcePopulation 73 stderr 8.709 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__rule_source_lookup/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate environmental esg](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__environmental_esg.md) | 84.59% | SE ±2.17 pp | 18 / 49 | $0.17 | 698.84source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.588 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.167209 \| USD \| per_test \| Lower is better \| \| `latency` \| 698.84 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.17 pp sourcePopulation 81 stderr 2.166 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__environmental_esg/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / data privacy cybersecurity](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Adata_privacy_cybersecurity.md) | 20% | SE ±20 pp | 4 / 49 | $0.18 | 561.46source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.178414 \| USD \| per_test \| Lower is better \| \| `latency` \| 561.456 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/data_privacy_cybersecurity/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / bpmn js](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Abpmn_js.md) | 60.38% | SE ±6.78 pp | 9 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 60.377 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.78 pp sourcePopulation 20 stderr 6.783 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/bpmn_js/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / trusts estates private client](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Atrusts_estates_private_client.md) | 20% | SE ±20 pp | 2 / 49 | $0.15 | 436.98source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.15293 \| USD \| per_test \| Lower is better \| \| `latency` \| 436.982 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/trusts_estates_private_client/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / capital markets](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acapital_markets.md) | 0% | SE ±0 pp | 16 / 49 | $0.16 | 456.47source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.16244 \| USD \| per_test \| Lower is better \| \| `latency` \| 456.474 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/capital_markets/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena vision style control / chinese](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Achinese.md) | 1,311.41 | — | 29 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1311.414012677891 \| rating \| reported \| Higher is better \| ### Configuration category chinese config vision_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1268.283773533959 rank 35 upper 1354.5442518218233 variance 484.24768178997476 votes 199 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/235/row` |  |  |  |  |  |  |  |
| [Mystery Game Puzzles](https://benchize.com/benchmarks/epoch-ai%3Amystery_game_puzzles.md) | 43% | SE ±4.98 pp | 14 / 71 | — | — | [epoch ai · Mystery Game Puzzles](https://epoch.ai/data/benchmark_data.zip#file=mystery_game_puzzles.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.43 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.43 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Mystery Game Puzzles reasoningBudget Not reported Reasoning effort max sourceModelVersion deepseek-v4.1-flash_max ### Reported statistics Uncertainty Standard error: ±4.98 pp standardError 0.0497569851956243 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=mystery_game_puzzles.csv) Source epoch ai Original model label deepseek-v4.1-flash Source version Mystery Game Puzzles Evaluated 2026-10-08T19:02:32.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `795b977bc82c460a5e2c1d6dd7b2569ddb21853f3829c4f6fdd16edfaf47bfcf` Source locator `/rows/3` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate.md) | 87.5% | SE ±0.4 pp | 19 / 49 | $0.17 | 579.44source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.495 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.173471 \| USD \| per_test \| Lower is better \| \| `latency` \| 579.44 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.4 pp sourcePopulation 81 stderr 0.398 tokenTotals {"cache_read_tokens":699334016,"cache_write_tokens":null,"input_tokens":721200533,"output_tokens":8383826,"reasoning_tokens":4048611} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass forms filings](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__forms_filings.md) | 56.78% | SE ±8.16 pp | 28 / 46 | $0.22 | 520.5source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 56.777 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.215479 \| USD \| per_test \| Lower is better \| \| `latency` \| 520.504 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.16 pp sourcePopulation 73 stderr 8.157 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__forms_filings/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass fact pattern analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__fact_pattern_analysis.md) | 60.62% | SE ±6.91 pp | 17 / 46 | $0.14 | 408.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 60.624 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.136181 \| USD \| per_test \| Lower is better \| \| `latency` \| 408.316 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.91 pp sourcePopulation 73 stderr 6.913 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__fact_pattern_analysis/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [BrokenArXiv Overall](https://benchize.com/benchmarks/matharena%3Aoverall--brokenarxiv.md) | 49.61% | 95% CI 44.2–55.02% | 3 / 4 | 0.3usd per problem | — | [matharena · overall--brokenarxiv](https://matharena.ai/competition_tables/overall--brokenarxiv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 49.61 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.30 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition overall--brokenarxiv predicted false Reasoning effort max sourceModelName DeepSeek-V4.1-Flash (Max) sourceModelVersion deepseek/deepseek_v41_flash thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 44.2–55.02% confidenceInterval95 5.41 problems Not reported ### Source [Original source](https://matharena.ai/competition_tables/overall--brokenarxiv) Source matharena Original model label DeepSeek-V4.1-Flash (Max) Source version overall--brokenarxiv Observed 2026-10-21T16:52:23.171Z Evidence SHA-256 `8af403d03bad2e0ac76660844e9b21df38cc214b99fc352f6374309a5ea83df6` Source locator `/table/rows/2` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass administrative regulatory](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__administrative_regulatory.md) | 89.06% | SE ±2.16 pp | 9 / 50 | $0.25 | 656.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.06 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.246492 \| USD \| per_test \| Lower is better \| \| `latency` \| 656.731 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 82 stderr 2.16 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__administrative_regulatory/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,500.99 | — | 23 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1500.987332092386 \| rating \| reported \| Higher is better \| ### Configuration category expert config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1485.7683095310367 rank 27 upper 1516.2063546537354 variance 60.29445024247478 votes 1496 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2497/row` |  |  |  |  |  |  |  |
| [ToneBench Scriptwriting](https://benchize.com/benchmarks/tonebench%3Ascripts.md) | 2,017.3 | Range 1,980–2,055.4 | 31 / 126 | — | — | [tonebench](https://benchmark.towardsai.com/data/leaderboard.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 2017.3 \| rating \| reported \| Higher is better \| \| `overall` \| 81.03 \| percent \| reported \| Higher is better \| \| `anti_slop_numbers` \| 71.02 \| source_score \| rubric \| Higher is better \| \| `continuity_emotion` \| 80.07 \| source_score \| rubric \| Higher is better \| \| `cue_quality` \| 75.98 \| source_score \| rubric \| Higher is better \| \| `hook_strength` \| 80.1 \| source_score \| rubric \| Higher is better \| \| `length_adherence` \| 95.94 \| source_score \| rubric \| Higher is better \| \| `slop` \| 83.17 \| source_score \| rubric \| Higher is better \| \| `slop_score` \| 95.33 \| source_score \| rubric \| Higher is better \| \| `substance_accuracy` \| 82.45 \| source_score \| rubric \| Higher is better \| \| `tone_voice` \| 78.12 \| source_score \| rubric \| Higher is better \| \| `writing_quality` \| 78.01 \| source_score \| rubric \| Higher is better \| \| `youtube_best_practices` \| 79.84 \| source_score \| rubric \| Higher is better \| ### Configuration provider openrouter Reasoning effort max scoreScope 10_task_current sourceConfiguration deepseek-v4-1-flash-max-or thinkingEnabled true ### Reported statistics Uncertainty Reported interval: 1,980–2,055.4 interval [1980,2055.4] rank 31 runCount 5 ### Source [Original source](https://benchmark.towardsai.com/data/leaderboard.json) Source tonebench Original model label DeepSeek V4.1 Flash (max) Evaluated 2026-10-08 Observed 2026-10-21T16:32:48.862Z Evidence SHA-256 `4264338337f418d3ded7ace250cad64fe12cacbd4b79a312bfaf0ae9268a01c3` Source locator `/rows/30` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / prettier](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Aprettier.md) | 61.54% | SE ±14.04 pp | 4 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 61.538 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±14.04 pp sourcePopulation 20 stderr 14.044 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/prettier/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Public Benefits Bench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Apublic-benefits-bench%3Aoverall.md) | 64.28% | SE ±1.25 pp | 15 / 34 | $0.07 | 598.96source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_snap.CUcEPb0v.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.276 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.073234 \| USD \| per_test \| Lower is better \| \| `latency` \| 598.96 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.25 pp sourcePopulation 50 stderr 1.246 tokenTotals {"cache_read_tokens":305366398,"cache_write_tokens":null,"input_tokens":324083542,"output_tokens":7830427,"reasoning_tokens":5092304} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_snap.CUcEPb0v.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1.1 Observed 2026-10-21T16:46:17.865Z Evidence SHA-256 `6f091831ab607a26d65d2ade18cba7ce345ea3a67297b8c378e8a06ce398360d` Source locator `/tasks/overall/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / all pass](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aall_pass.md) | 51.71% | SE ±2.21 pp | 7 / 18 | $0.37 | 960.25source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 51.708 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.366574 \| USD \| per_test \| Lower is better \| \| `latency` \| 960.248 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.21 pp sourcePopulation 28 stderr 2.209 tokenTotals {"cache_read_tokens":389922803,"cache_write_tokens":null,"input_tokens":596577990,"output_tokens":83546431,"reasoning_tokens":74653360} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/all_pass/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / insurance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainsurance.md) | 0% | SE ±0 pp | 2 / 49 | $0.15 | 848.14source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.151326 \| USD \| per_test \| Lower is better \| \| `latency` \| 848.144 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/insurance/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate trusts estates private client](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__trusts_estates_private_client.md) | 88.17% | SE ±2 pp | 14 / 49 | $0.15 | 436.98source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.168 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.15293 \| USD \| per_test \| Lower is better \| \| `latency` \| 436.982 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2 pp sourcePopulation 81 stderr 1.999 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__trusts_estates_private_client/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text style control / polish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Apolish.md) | 1,472.2 | — | 26 / 117 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1472.2005522845932 \| rating \| reported \| Higher is better \| ### Configuration category polish config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1434.3653433973989 rank 39 upper 1510.0357611717877 variance 372.6456792471809 votes 222 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10186/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate data privacy cybersecurity](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__data_privacy_cybersecurity.md) | 93.99% | SE ±1.42 pp | 15 / 49 | $0.18 | 561.46source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.993 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.178414 \| USD \| per_test \| Lower is better \| \| `latency` \| 561.456 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.42 pp sourcePopulation 81 stderr 1.415 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__data_privacy_cybersecurity/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate real estate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__real_estate.md) | 90.67% | SE ±1.57 pp | 22 / 49 | $0.19 | 417.46source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.671 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.193926 \| USD \| per_test \| Lower is better \| \| `latency` \| 417.46 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.57 pp sourcePopulation 81 stderr 1.573 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__real_estate/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · earth](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aearth.md) | 8.33% | SE ±5.64 pp | 4 / 9 | $39.17 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 8.333333333333332 \| percent \| reported \| Higher is better \| \| `total_cost` \| 39.16776078 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 1777639333 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Codex benchmarkVersion v0-1-eval domain earth Reasoning effort max ### Reported statistics Uncertainty Standard error: ±5.64 pp accuracyStandardError 5.64169333655275 passes 2 rank 7 tasks 24 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label DeepSeek V4.1 Flash Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/6/metrics/domain_metrics/earth` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / employment labor](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aemployment_labor.md) | 0% | SE ±0 pp | 1 / 49 | $0.13 | 449.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.13211 \| USD \| per_test \| Lower is better \| \| `latency` \| 449.11 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/employment_labor/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena vision / homework (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Ahomework--config-vision.md) | 1,306.7 | — | 31 / 73 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1306.6980316785766 \| rating \| reported \| Higher is better \| ### Configuration category homework config vision Reasoning effort max thinkingEnabled true ### Reported statistics lower 1281.2389568688354 rank 34 upper 1332.1571064883178 variance 168.7287357282877 votes 521 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/841/row` |  |  |  |  |  |  |  |
| [ARC-AGI-3](https://benchize.com/benchmarks/arc-prize%3Av3_Semi_Private.md) | 0.85% | — | 26 / 53 | 1,164.91usd | — | [arc prize · v3](https://arcprize.org/media/data/leaderboard/v3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.008516646657220222 \| score_fraction \| v3_Semi_Private \| Higher is better \| \| `total_cost` \| 1164.9106122 \| usd \| v3_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v3_Semi_Private harness standard modelGroup deepseek-v4-1-flash modelType CoT Reasoning effort max sourceModelVersion deepseek-v4-1-flash-max thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v3.json) Source arc prize Original model label DeepSeek V4.1 Flash (Max) Source version v3 Observed 2026-10-21T16:34:21.112Z Evidence SHA-256 `1aaee2199f8c6d9b5a589aab2fd2480a729e9132f8cf8fbf7b49f81dbd5b2c49` Source locator `/evaluations/87` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass family](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__family.md) | 77.73% | SE ±4.7 pp | 8 / 50 | $0.32 | 1,013.4source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.732 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.323772 \| USD \| per_test \| Lower is better \| \| `latency` \| 1013.398 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.7 pp sourcePopulation 82 stderr 4.697 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__family/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text style control / english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aenglish.md) | 1,486.24 | — | 25 / 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` \| 1486.2418563945266 \| rating \| reported \| Higher is better \| ### Configuration category english config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1477.461071202138 rank 28 upper 1495.022641586915 variance 20.07106992258094 votes 4947 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1664/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate energy natural resources](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__energy_natural_resources.md) | 93.07% | SE ±1.46 pp | 24 / 49 | $0.22 | 398.4source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.069 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.219636 \| USD \| per_test \| Lower is better \| \| `latency` \| 398.398 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.46 pp sourcePopulation 81 stderr 1.461 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__energy_natural_resources/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · engineering](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aengineering.md) | 18.52% | SE ±7.48 pp | 2 / 9 | $69.75 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 18.51851851851852 \| percent \| reported \| Higher is better \| \| `total_cost` \| 69.74610236399998 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 3041093185 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Codex benchmarkVersion v0-1-eval domain engineering Reasoning effort max ### Reported statistics Uncertainty Standard error: ±7.48 pp accuracyStandardError 7.4756798866604734 passes 5 rank 7 tasks 27 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label DeepSeek V4.1 Flash Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/6/metrics/domain_metrics/engineering` |  |  |  |  |  |  |  |
| [ProgramBench / partial](https://benchize.com/benchmarks/vals-ai%3Aprogrambench%3Apartial.md) | 72% | SE ±1.99 pp | 14 / 40 | $0.9 | 3,709.71source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 71.999 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.896935 \| USD \| per_test \| Lower is better \| \| `latency` \| 3709.707 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.99 pp sourcePopulation 64 stderr 1.987 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:10.137Z Evidence SHA-256 `e9bbb4378c7a058441a21d456baf9316b96c2cf5b58d371a3c7c803f5000bd8c` Source locator `/tasks/partial/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate antitrust competition](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__antitrust_competition.md) | 87.68% | SE ±1.98 pp | 17 / 49 | $0.17 | 746.25source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.681 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.170347 \| USD \| per_test \| Lower is better \| \| `latency` \| 746.252 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.98 pp sourcePopulation 81 stderr 1.982 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__antitrust_competition/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate tax](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__tax.md) | 82.27% | SE ±2.21 pp | 19 / 49 | $0.2 | 502.12source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.274 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.198778 \| USD \| per_test \| Lower is better \| \| `latency` \| 502.124 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.21 pp sourcePopulation 81 stderr 2.212 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__tax/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / family](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Afamily.md) | 13.64% | SE ±7.49 pp | 28 / 50 | $0.32 | 1,013.4source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 13.636 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.323772 \| USD \| per_test \| Lower is better \| \| `latency` \| 1013.398 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.49 pp sourcePopulation 82 stderr 7.489 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/family/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / healthcare life sciences](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ahealthcare_life_sciences.md) | 20% | SE ±20 pp | 5 / 49 | $0.17 | 622.19source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.166133 \| USD \| per_test \| Lower is better \| \| `latency` \| 622.19 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/healthcare_life_sciences/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / funds asset management](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Afunds_asset_management.md) | 0% | SE ±0 pp | 11 / 49 | $0.19 | 828.74source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.190615 \| USD \| per_test \| Lower is better \| \| `latency` \| 828.736 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/funds_asset_management/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate ma](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_ma.md) | 91.99% | SE ±1.54 pp | 19 / 49 | $0.17 | 444.37source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.987 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.166084 \| USD \| per_test \| Lower is better \| \| `latency` \| 444.366 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.54 pp sourcePopulation 81 stderr 1.539 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_ma/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [BullshitBench v2](https://benchize.com/benchmarks/bullshitbench%3Av2.md) | 57% | — | 53 / 191 | — | — | [bullshitbench · v2](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `green_rate` \| 0.57 \| score_fraction \| all_attempts \| Higher is better \| \| `red_rate` \| 0.13 \| score_fraction \| all_attempts \| Lower is better \| \| `refusal_rate` \| 0.0 \| score_fraction \| all_attempts \| Lower is better \| \| `avg_score` \| 1.4167 \| score \| judge_grade_0_to_2 \| Higher is better \| ### Configuration Reasoning effort max suite v2 thinkingEnabled true ### Reported statistics acceptedCount 13 answeredCount 100 clearCount 57 errorCount 0 partialCount 30 rank 63 refusalCount 0 sampleSize 100 ### Source [Original source](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) Source bullshitbench Original model label deepseek/deepseek-v4.1-flash Source version v2 Observed 2026-10-21T16:33:56.675Z Evidence SHA-256 `98b485e08764d1c97144bc4f7b28ced1a96d50850947c775870dfda0e5d749b5` Source locator `/rows/62` |  |  |  |  |  |  |  |
| [Arena text style control / creative writing](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acreative_writing.md) | 1,440.96 | — | 38 / 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` \| 1440.9579675584378 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1428.912250072187 rank 56 upper 1453.0036850446882 variance 37.77192898096693 votes 2759 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1277/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) | 1,449.7 | — | 29 / 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` \| 1449.7024647198994 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1438.6561862892322 rank 34 upper 1460.7487431505665 variance 31.76404404245927 votes 3327 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4726/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate intellectual property](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__intellectual_property.md) | 91.52% | SE ±1.66 pp | 17 / 49 | $0.17 | 564.49source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.519 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.166991 \| USD \| per_test \| Lower is better \| \| `latency` \| 564.488 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.66 pp sourcePopulation 81 stderr 1.659 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__intellectual_property/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate white collar defense investigations](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__white_collar_defense_investigations.md) | 86.2% | SE ±2.01 pp | 14 / 49 | $0.15 | 772.2source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.195 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.146247 \| USD \| per_test \| Lower is better \| \| `latency` \| 772.198 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.01 pp sourcePopulation 81 stderr 2.005 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__white_collar_defense_investigations/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__business_commercial.md) | 86.68% | SE ±2.42 pp | 22 / 50 | $0.23 | 694.59source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.675 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.226487 \| USD \| per_test \| Lower is better \| \| `latency` \| 694.588 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.42 pp sourcePopulation 82 stderr 2.424 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__business_commercial/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena webdev / overall](https://benchize.com/benchmarks/lmarena%3Awebdev%3Aoverall.md) | 1,618.34 | — | 20 / 90 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1618.3443578289887 \| rating \| reported \| Higher is better \| ### Configuration category overall config webdev Reasoning effort max thinkingEnabled true ### Reported statistics lower 1607.388798400928 rank 26 upper 1629.2999172570492 variance 31.244453730752724 votes 3528 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/25/row` |  |  |  |  |  |  |  |
| [Arena text / russian](https://benchize.com/benchmarks/lmarena%3Atext%3Arussian.md) | 1,467.06 | — | 30 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1467.062388788605 \| rating \| reported \| Higher is better \| ### Configuration category russian config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1453.304399565397 rank 35 upper 1480.8203780118129 variance 49.27353807523117 votes 1920 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10418/row` |  |  |  |  |  |  |  |
| [Arena vision style control / diagram](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Adiagram.md) | 1,293.37 | — | 35 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1293.368350333374 \| rating \| reported \| Higher is better \| ### Configuration category diagram config vision_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1274.6917238327555 rank 41 upper 1312.0449768339925 variance 90.80310208312302 votes 1085 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/507/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / tax](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Atax.md) | 0% | SE ±0 pp | 3 / 49 | $0.2 | 502.12source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.198778 \| USD \| per_test \| Lower is better \| \| `latency` \| 502.124 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/tax/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / white collar defense investigations](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Awhite_collar_defense_investigations.md) | 0% | SE ±0 pp | 1 / 49 | $0.15 | 772.2source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.146247 \| USD \| per_test \| Lower is better \| \| `latency` \| 772.198 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/white_collar_defense_investigations/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__international_trade_sanctions.md) | 86.91% | SE ±2.13 pp | 19 / 49 | $0.16 | 555.68source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.905 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.163871 \| USD \| per_test \| Lower is better \| \| `latency` \| 555.678 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.13 pp sourcePopulation 81 stderr 2.129 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__international_trade_sanctions/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [TapTap Maker heldout](https://benchize.com/benchmarks/taptap-maker%3Aheldout.md) | 59.41% | Range 59.41–59.41% | 14 / 26 | — | — | [taptap maker](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `l2_pass_rate` \| 0.5941 \| score_fraction \| reported \| Higher is better \| ### Configuration datasetVersion 0.2.3 edition 2026-10-18-v6-voxel harness claude Reasoning effort max reasoningMode Not reported ### Reported statistics Uncertainty Reported interval: 59.41–59.41% interval [0.5941,0.5941] intervalMethod bootstrap-case rank 16 rankInterval [14,17] ### Source [Original source](https://maker.taptap.cn/leaderboard/data/editions/2026-10-18-v6-voxel.json) Source taptap maker Original model label deepseek-v4.1-flash Observed 2026-10-21T16:33:19.620Z Evidence SHA-256 `ad2d7f1adeae0548a4edc0b5e75d11449b443449ff76533d04bc1fb1b6d7035e` Source locator `/heldout/15` |  |  |  |  |  |  |  |
| [Arena vision style control / english](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Aenglish.md) | 1,268.14 | — | 33 / 82 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1268.1413452888005 \| rating \| reported \| Higher is better \| ### Configuration category english config vision_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1250.8072408966348 rank 39 upper 1285.4754496809664 variance 78.21798673458414 votes 1301 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/628/row` |  |  |  |  |  |  |  |
| [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,483.69 | — | 19 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1483.6863012824083 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1463.617433347547 rank 22 upper 1503.7551692172692 variance 104.84544517754665 votes 910 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6334/row` |  |  |  |  |  |  |  |
| [Arena vision style control / creative writing vision](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Acreative_writing_vision.md) | 1,254.27 | — | 35 / 70 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1254.269268641075 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing_vision config vision_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1220.8327824409207 rank 42 upper 1287.7057548412295 variance 291.03490668451286 votes 345 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/404/row` |  |  |  |  |  |  |  |
| [Arena text style control / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_legal_and_government.md) | 1,456.36 | — | 54 / 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` \| 1456.3608327697862 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1438.70614922905 rank 86 upper 1474.0155163105221 variance 81.13788679562519 votes 1145 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5193/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass immigration](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__immigration.md) | 77.31% | SE ±9.06 pp | 32 / 50 | $0.44 | 1,059.33source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.311 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.440884 \| USD \| per_test \| Lower is better \| \| `latency` \| 1059.33 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.06 pp sourcePopulation 82 stderr 9.055 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__immigration/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass civil litigation](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__civil_litigation.md) | 87.1% | SE ±3.71 pp | 25 / 50 | $0.24 | 663.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.099 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.239442 \| USD \| per_test \| Lower is better \| \| `latency` \| 663.732 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.71 pp sourcePopulation 82 stderr 3.706 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__civil_litigation/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aoverall.md) | 15.71% | SE ±2.51 pp | 3 / 9 | $385.93 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 15.714285714285714 \| percent \| reported \| Higher is better \| \| `total_cost` \| 385.933474254 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 14859405919 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Codex benchmarkVersion v0-1-eval domain overall Reasoning effort max ### Reported statistics Uncertainty Standard error: ±2.51 pp accuracyStandardError 2.511392893650442 passes 33 rank 7 tasks 210 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label DeepSeek V4.1 Flash Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/6/metrics` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate governance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_governance.md) | 91.44% | SE ±1.64 pp | 11 / 49 | $0.18 | 676.24source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.438 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.182687 \| USD \| per_test \| Lower is better \| \| `latency` \| 676.244 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.64 pp sourcePopulation 81 stderr 1.64 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_governance/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate employment labor](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__employment_labor.md) | 82.02% | SE ±2.35 pp | 17 / 49 | $0.13 | 449.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.022 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.13211 \| USD \| per_test \| Lower is better \| \| `latency` \| 449.11 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.35 pp sourcePopulation 81 stderr 2.354 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__employment_labor/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / general qualitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Ageneral_qualitative_analysis.md) | 72.99% | SE ±5.43 pp | 6 / 18 | $0.4 | 1,087.08source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.992 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.401506 \| USD \| per_test \| Lower is better \| \| `latency` \| 1087.084 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.43 pp sourcePopulation 28 stderr 5.431 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/general_qualitative_analysis/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / general quantitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Ageneral_quantitative_analysis.md) | 76.52% | SE ±5.4 pp | 12 / 18 | $0.33 | 734.76source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 76.517 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.333817 \| USD \| per_test \| Lower is better \| \| `latency` \| 734.763 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.4 pp sourcePopulation 28 stderr 5.398 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/general_quantitative_analysis/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [APEX-Agents](https://benchize.com/benchmarks/mercor%3Aapex-agents%3A1.1.md) | 39.5% | — | 29 / 38 | — | — | [mercor · 1.1](https://www.mercor.com/apex/apex-agents-leaderboard/) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 39.5 \| percent \| task_completion_pass_at_1 \| Higher is better \| \| `mean_score` \| 50.5 \| percent \| rubric_mean \| Higher is better \| ### Configuration datasetUrl https://huggingface.co/datasets/mercor/apex-agents-v1.1 datasetVersion 1.1 harness loop_truncated_tools_agent Reasoning effort max ### Reported statistics reportedErrors {"mean_score":4.7,"pass_1":4.9} sampleSize 960 ### Source [Original source](https://www.mercor.com/apex/apex-agents-leaderboard/) Source mercor Original model label DeepSeek-V4.1-Flash Source version 1.1 Observed 2026-10-21T16:33:27.509Z Evidence SHA-256 `359c569c3b0b62ab375895eabf3c45846fa45551d419d5c9d6a4c9117db78184` Source locator `/__NEXT_DATA__/props/pageProps/leaderboardData/44` |  |  |  |  |  |  |  |
| [Arena text style control / expert](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexpert.md) | 1,515.72 | — | 19 / 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` \| 1515.718005633411 \| rating \| reported \| Higher is better \| ### Configuration category expert config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1500.4521164943665 rank 21 upper 1530.9838947724556 variance 60.66637235577339 votes 1496 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2491/row` |  |  |  |  |  |  |  |
| [Arena text / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts_english.md) | 1,488.04 | — | 20 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1488.0397931459825 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1477.4361475404057 rank 23 upper 1498.6434387515587 variance 29.269427417249716 votes 3257 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3890/row` |  |  |  |  |  |  |  |
| [Tax Agent Bench / current temporal analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Acurrent_temporal_analysis.md) | 39% | SE ±8.92 pp | 17 / 46 | $0.15 | 397.46source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 39 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.145689 \| USD \| per_test \| Lower is better \| \| `latency` \| 397.458 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.92 pp sourcePopulation 73 stderr 8.923 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/current_temporal_analysis/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass.md) | 62.46% | SE ±3.16 pp | 20 / 46 | $0.13 | 349.08source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 62.458 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.134662 \| USD \| per_test \| Lower is better \| \| `latency` \| 349.076 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.16 pp sourcePopulation 73 stderr 3.155 tokenTotals {"cache_read_tokens":91970688,"cache_write_tokens":null,"input_tokens":114496920,"output_tokens":15566679,"reasoning_tokens":12729723} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [EMB v1.1 / dataroom](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Adataroom.md) | 79.94% | SE ±5.96 pp | 17 / 19 | $0.2 | 695.25source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.935 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.196161 \| USD \| per_test \| Lower is better \| \| `latency` \| 695.253 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.96 pp sourcePopulation 30 stderr 5.964 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/dataroom/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [ProgramBench / strict](https://benchize.com/benchmarks/vals-ai%3Aprogrambench%3Astrict.md) | 0.5% | SE ±0.5 pp | 14 / 40 | $0.9 | 3,709.71source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0.5 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.896935 \| USD \| per_test \| Lower is better \| \| `latency` \| 3709.707 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.5 pp sourcePopulation 64 stderr 0.5 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:10.137Z Evidence SHA-256 `e9bbb4378c7a058441a21d456baf9316b96c2cf5b58d371a3c7c803f5000bd8c` Source locator `/tasks/strict/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / corporate governance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acorporate_governance.md) | 0% | SE ±0 pp | 11 / 49 | $0.18 | 676.24source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.182687 \| USD \| per_test \| Lower is better \| \| `latency` \| 676.244 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/corporate_governance/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena webdev / webdev-react](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev-react.md) | 1,628.47 | — | 20 / 81 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1628.4736628645387 \| rating \| reported \| Higher is better \| ### Configuration category webdev-react config webdev Reasoning effort max thinkingEnabled true ### Reported statistics lower 1614.5161568527305 rank 27 upper 1642.4311688763469 variance 50.71301897607303 votes 2162 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/533/row` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / carbon](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Acarbon.md) | 29.32% | SE ±3.96 pp | 5 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 29.323 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.96 pp sourcePopulation 20 stderr 3.962 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/carbon/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Abusiness_commercial.md) | 39.54% | SE ±7.54 pp | 15 / 50 | $0.23 | 694.59source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 39.535 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.226487 \| USD \| per_test \| Lower is better \| \| `latency` \| 694.588 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.54 pp sourcePopulation 82 stderr 7.544 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/business_commercial/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [Arena text / multi turn](https://benchize.com/benchmarks/lmarena%3Atext%3Amulti_turn.md) | 1,460.78 | — | 36 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1460.7788183735302 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1447.1236982277994 rank 47 upper 1474.433938519261 variance 48.5394520409432 votes 1944 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9362/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / intellectual property](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aintellectual_property.md) | 0% | SE ±0 pp | 6 / 49 | $0.17 | 564.49source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.166991 \| USD \| per_test \| Lower is better \| \| `latency` \| 564.488 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/intellectual_property/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [EMB v1.1 / ma](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Ama.md) | 75.71% | SE ±4.75 pp | 10 / 19 | $0.19 | 661.5source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 75.714 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.194981 \| USD \| per_test \| Lower is better \| \| `latency` \| 661.499 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.75 pp sourcePopulation 30 stderr 4.754 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/ma/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |  |
| [ArXivLean 09/2026](https://benchize.com/benchmarks/matharena%3Aarxivlean--september.md) | 5.66% | 95% CI -0.56–11.88% | 5 / 6 | 1.87usd per problem | — | [matharena · arxivlean--september](https://matharena.ai/competition_tables/arxivlean--september) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 5.66 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 1.87 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition arxivlean--september predicted false Reasoning effort max sourceModelName DeepSeek-V4.1-Flash (Max) sourceModelVersion deepseek/deepseek_v41_flash thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -0.56–11.88% confidenceInterval95 6.22 problems 53 ### Source [Original source](https://matharena.ai/competition_tables/arxivlean--september) Source matharena Original model label DeepSeek-V4.1-Flash (Max) Source version arxivlean--september Observed 2026-10-21T16:52:22.825Z Evidence SHA-256 `5d42ba0bbc37cc684861959960e17e45edb8cd93395dfe74b44e3252f043c42d` Source locator `/table/rows/4` |  |  |  |  |  |  |  |
| [BrokenArXiv 05/2026](https://benchize.com/benchmarks/matharena%3Aarxiv_false--may.md) | 28.5% | 95% CI 19.65–37.35% | 10 / 20 | 0.03usd per problem | — | [matharena · arxiv_false--may](https://matharena.ai/competition_tables/arxiv_false--may) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 28.50 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.033 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition arxiv_false--may predicted false Reasoning effort max sourceModelName DeepSeek-V4.1-Flash (Max) sourceModelVersion deepseek/deepseek_v41_flash thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 19.65–37.35% confidenceInterval95 8.85 problems 50 ### Source [Original source](https://matharena.ai/competition_tables/arxiv_false--may) Source matharena Original model label DeepSeek-V4.1-Flash (Max) Source version arxiv_false--may Observed 2026-10-21T16:52:24.785Z Evidence SHA-256 `b317fca30109e8c78ed5d2b835fe227b8cfbee3b6f9cb6d7aaeadca866660969` Source locator `/table/rows/12` |  |  |  |  |  |  |  |
| [Arena text style control / french](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Afrench.md) | 1,480.22 | — | 40 / 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` \| 1480.2218353109483 \| rating \| reported \| Higher is better \| ### Configuration category french config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1453.0592791375639 rank 61 upper 1507.3843914843328 variance 192.06361236978 votes 520 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label DeepSeek-V4.1-Flash-20260910 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2898/row` |  |  |  |  |  |  |  |

[Next page](https://benchize.com/models/deepseek-v4-1-flash-max.md?cursor=614407c0d1b20cb9481cb3df29cb784ac999be0ed772e4722dbfadd10bdfe91e)

---
Cite as: DeepSeek V4.1 FlashMax — Benchize, https://benchize.com/models/deepseek-v4-1-flash-max (retrieved 2026-10-21)