# Gemini 3.7 FlashHigh

> Source: Benchize (https://benchize.com) — https://benchize.com/models/gemini-3-7-flash-high. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gemini-3-7-flash-high.

Google

Canonical URL: https://benchize.com/models/gemini-3-7-flash-high
Last updated: Wed, 21 Oct 2026 18:05:10 GMT

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

Google

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [SWE-bench / >4 hours](https://benchize.com/benchmarks/vals-ai%3Aswebench%3A%3E4%20hours.md) | 33.33% | SE ±33.33 pp | 13 / 39 | $2.75 | 634.86source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.333 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.747013 \| USD \| per_test \| Lower is better \| \| `latency` \| 634.863 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±33.33 pp sourcePopulation 88 stderr 33.333 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:13.711Z Evidence SHA-256 `7a4e2e8485283caac500127419f68ec4bd5414c837ade114fc6afa5b2d34ccaa` Source locator `/tasks/>4 hours/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [FrontierMath Tier 4 v2 (private)](https://benchize.com/benchmarks/epoch-ai%3Afrontiermath_tier_4_v2.md) | 36.59% | SE ±7.62 pp | 24 / 38 | — | — | [epoch ai · FrontierMath-Tier-4-v2-Private](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tier_4_v2.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.36585365853658536 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.36585365853658536 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion FrontierMath-Tier-4-v2-Private reasoningBudget Not reported Reasoning effort high sourceModelVersion gemini-3.7-flash_high ### Reported statistics Uncertainty Standard error: ±7.62 pp standardError 0.07615851217559022 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tier_4_v2.csv) Source epoch ai Original model label gemini-3.7-flash Source version FrontierMath-Tier-4-v2-Private Evaluated 2026-08-14T19:59:58.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `48b886c828092231989fba6e3819efb7462e89ef681b7b306b4bfa3505829e4a` Source locator `/rows/26` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / emerging companies venture capital](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aemerging_companies_venture_capital.md) | 0% | SE ±0 pp | 18 / 49 | $2.02 | 406.88source 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` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.01759 \| USD \| per_test \| Lower is better \| \| `latency` \| 406.88 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/emerging_companies_venture_capital/google~1gemini-3.7-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) | 79.03% | SE ±2.5 pp | 24 / 49 | $2.29 | 704.45source 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` \| 79.026 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.288046 \| USD \| per_test \| Lower is better \| \| `latency` \| 704.448 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.5 pp sourcePopulation 81 stderr 2.496 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__employment_labor/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Furniture Assembly Benchmark](https://benchize.com/benchmarks/epoch-ai%3Afurniture_assembly.md) | 26.67% | SE ±5.63 pp | 18 / 21 | — | — | [epoch ai · Furniture Assembly](https://epoch.ai/data/benchmark_data.zip#file=furniture_assembly.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.26666666666666666 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.26666666666666666 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Furniture Assembly reasoningBudget Not reported Reasoning effort high sourceModelVersion gemini-3.7-flash_high ### Reported statistics Uncertainty Standard error: ±5.63 pp standardError 0.056331678510110804 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=furniture_assembly.csv) Source epoch ai Original model label gemini-3.7-flash Source version Furniture Assembly Evaluated 2026-09-10T20:23:57.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `d62ab28005a1b7ab0e57b227d04b0cf86c70a5d8a1981addb26d2ac41c116865` Source locator `/rows/16` |  |  |  |  |  |  |  |
| [Short Story Writing — Pairwise Comparison](https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise.md) | -0.71 | 95% CI -0.83–-0.6 | 38 / 50 | — | — | [lechmazur short story · 8ba548d1cffefa5c472d3832e05f64e119a88fa4](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| -0.7060157386399251 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.3927325750567854 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort high thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: -0.83–-0.6 bootstrapSamples 300 confidenceInterval95 ["-0.8319882408203993","-0.6015755844436229"] sourceRank 39 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label Gemini 3.7 Flash (high) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/38` |  |  |  |  |  |  |  |
| [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,493.86 | — | 14 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1493.8581232416154 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1480.2746796641773 rank 15 upper 1507.4415668190532 variance 48.031216273223 votes 2096 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6327/row` |  |  |  |  |  |  |  |
| [Arena text style control / coding](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acoding.md) | 1,518.7 | — | 32 / 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` \| 1518.7010928907646 \| rating \| reported \| Higher is better \| ### Configuration category coding config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1511.1430497557171 rank 40 upper 1526.259136025812 variance 14.870396559637385 votes 7905 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/849/row` |  |  |  |  |  |  |  |
| [IOI / ioi2026](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aioi2026.md) | 60.83% | SE ±16.63 pp | 11 / 29 | $5.39 | 2,475.47source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 60.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.392624 \| USD \| per_test \| Lower is better \| \| `latency` \| 2475.465 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±16.63 pp sourcePopulation 44 stderr 16.627 tokenTotals {"cache_read_tokens":60262017,"cache_write_tokens":0,"input_tokens":70967304,"output_tokens":967801,"reasoning_tokens":429792} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/ioi2026/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / bankruptcy restructuring](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Abankruptcy_restructuring.md) | 0% | SE ±0 pp | 3 / 49 | $3.53 | 358.7source 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` \| 3.528761 \| USD \| per_test \| Lower is better \| \| `latency` \| 358.702 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/bankruptcy_restructuring/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate structured finance securitization](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__structured_finance_securitization.md) | 86.98% | SE ±1.9 pp | 8 / 49 | $2.18 | 678.65source 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` \| 86.984 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.18141 \| USD \| per_test \| Lower is better \| \| `latency` \| 678.652 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.9 pp sourcePopulation 81 stderr 1.899 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__structured_finance_securitization/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena vision style control / homework](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Ahomework.md) | 1,308.9 | — | 23 / 73 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1308.9002020990436 \| rating \| reported \| Higher is better \| ### Configuration category homework config vision_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1286.1809161086283 rank 24 upper 1331.619488089459 variance 134.36717143333004 votes 665 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/831/row` |  |  |  |  |  |  |  |
| [Arena text style control / polish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Apolish.md) | 1,499.38 | — | 10 / 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` \| 1499.3759652937265 \| rating \| reported \| Higher is better \| ### Configuration category polish config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1471.4731485531336 rank 10 upper 1527.2787820343194 variance 202.67487389554566 votes 453 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10157/row` |  |  |  |  |  |  |  |
| [Arena text style control / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts_english.md) | 1,497.17 | — | 32 / 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` \| 1497.1722447872485 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1489.5125757124747 rank 41 upper 1504.8319138620227 variance 15.272981716994265 votes 7632 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3908/row` |  |  |  |  |  |  |  |
| [Arena vision style control / english](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Aenglish.md) | 1,287.36 | — | 9 / 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` \| 1287.3644721640032 \| rating \| reported \| Higher is better \| ### Configuration category english config vision_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1272.5756424083163 rank 10 upper 1302.1533019196904 variance 56.93396591016274 votes 1848 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/599/row` |  |  |  |  |  |  |  |
| [Terminal-Bench 4](https://benchize.com/benchmarks/harbor%3Aterminal-bench-4.md) | 11.21% | 95% CI 8.76–13.66% | 17 / 17 | — | — | [harbor](https://www.tbench.ai/) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 11.21 \| percent \| reported \| Higher is better \| \| `pass_at_2` \| 0.1636 \| score_fraction \| reported \| Higher is better \| \| `pass_at_3` \| 0.2 \| score_fraction \| reported \| Higher is better \| \| `pass_at_4` \| 0.2303 \| score_fraction \| reported \| Higher is better \| \| `pass_at_5` \| 0.2576 \| score_fraction \| reported \| Higher is better \| ### Configuration agent mini-SWE-agent Reasoning effort high sourceSubmission 14f4da14-b86a-4dca-92d7-11178d4fe064 ### Reported statistics Uncertainty 95% confidence interval: 8.76–13.66% confidenceHalfWidth95 2.45 sampleSize 330 ### Source [Original source](https://www.tbench.ai/) Source harbor Original model label Gemini 3.7 Flash Evaluated 2026-08-13 Observed 2026-10-21T16:33:12.399Z Evidence SHA-256 `0c2d7608e0711b34acae3e757dc2b8bf58bf1f5f04fec84de6cd02598168712c` Source locator `/rows/34` |  |  |  |  |  |  |  |
| [SWE-bench / overall](https://benchize.com/benchmarks/vals-ai%3Aswebench%3Aoverall.md) | 80.8% | SE ±1.76 pp | 13 / 39 | $1.44 | 350.44source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.8 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.436459 \| USD \| per_test \| Lower is better \| \| `latency` \| 350.439 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.76 pp sourcePopulation 88 stderr 1.763 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:13.711Z Evidence SHA-256 `7a4e2e8485283caac500127419f68ec4bd5414c837ade114fc6afa5b2d34ccaa` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [MMLU Pro / psychology](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Apsychology.md) | 91.1% | SE ±1.01 pp | 7 / 40 | $0.0067 | 2.92source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.103 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.006702 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.915 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.01 pp sourcePopulation 138 stderr 1.008 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/psychology/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / adjustments](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aadjustments.md) | 54.72% | SE ±0.74 pp | 7 / 50 | $1.51 | 171.63source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 54.718 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.506148 \| USD \| per_test \| Lower is better \| \| `latency` \| 171.627 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.74 pp sourcePopulation 80 stderr 0.739 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/adjustments/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / instruction following](https://benchize.com/benchmarks/lmarena%3Atext%3Ainstruction_following.md) | 1,484.45 | — | 15 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1484.4526261840829 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1477.5768728487353 rank 16 upper 1491.3283795194302 variance 12.306778787751721 votes 10486 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7549/row` |  |  |  |  |  |  |  |
| [ToneBench Scriptwriting](https://benchize.com/benchmarks/tonebench%3Ascripts.md) | 1,153.6 | Range 1,098.8–1,205.5 | 92 / 126 | — | — | [tonebench](https://benchmark.towardsai.com/data/leaderboard.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 1153.6 \| rating \| reported \| Higher is better \| \| `overall` \| 70.06 \| percent \| reported \| Higher is better \| \| `anti_slop_numbers` \| 60.78 \| source_score \| rubric \| Higher is better \| \| `continuity_emotion` \| 73.28 \| source_score \| rubric \| Higher is better \| \| `cue_quality` \| 74.23 \| source_score \| rubric \| Higher is better \| \| `hook_strength` \| 73.63 \| source_score \| rubric \| Higher is better \| \| `length_adherence` \| 68.59 \| source_score \| rubric \| Higher is better \| \| `slop` \| 75.57 \| source_score \| rubric \| Higher is better \| \| `slop_score` \| 90.36 \| source_score \| rubric \| Higher is better \| \| `substance_accuracy` \| 73.05 \| source_score \| rubric \| Higher is better \| \| `tone_voice` \| 60.87 \| source_score \| rubric \| Higher is better \| \| `writing_quality` \| 68.19 \| source_score \| rubric \| Higher is better \| \| `youtube_best_practices` \| 73.51 \| source_score \| rubric \| Higher is better \| ### Configuration provider google Reasoning effort high scoreScope 10_task_current sourceConfiguration gemini-3-7-flash-high ### Reported statistics Uncertainty Reported interval: 1,098.8–1,205.5 interval [1098.8,1205.5] rank 95 runCount 5 ### Source [Original source](https://benchmark.towardsai.com/data/leaderboard.json) Source tonebench Original model label Gemini 3.7 Flash (high thinking) Evaluated 2026-10-11 Observed 2026-10-21T16:32:48.862Z Evidence SHA-256 `4264338337f418d3ded7ace250cad64fe12cacbd4b79a312bfaf0ae9268a01c3` Source locator `/rows/94` |  |  |  |  |  |  |  |
| [Arena vision / humor (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Ahumor--config-vision.md) | 1,299.38 | — | 13 / 63 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1299.3830404298008 \| rating \| reported \| Higher is better \| ### Configuration category humor config vision Reasoning effort high thinkingEnabled true ### Reported statistics lower 1257.923991225286 rank 14 upper 1340.8420896343152 variance 447.4479204834404 votes 215 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/939/row` |  |  |  |  |  |  |  |
| [GPQA Diamond / diamond zero shot cot](https://benchize.com/benchmarks/vals-ai%3Agpqa%3Adiamond_zero_shot_cot.md) | 92.42% | SE ±2.4 pp | 13 / 40 | $0.02 | 8.36source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_gpqa.ghZZBpyU.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.424 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.02089 \| USD \| per_test \| Lower is better \| \| `latency` \| 8.36 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.4 pp sourcePopulation 138 stderr 2.4 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_gpqa.ghZZBpyU.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:06.040Z Evidence SHA-256 `951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077` Source locator `/tasks/diamond_zero_shot_cot/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [EMB / ma](https://benchize.com/benchmarks/vals-ai%3Aemb%3Ama.md) | 83.36% | SE ±3.22 pp | 3 / 47 | $3.87 | 718.87source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 83.364 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.868043 \| USD \| per_test \| Lower is better \| \| `latency` \| 718.865 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.22 pp sourcePopulation 76 stderr 3.22 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/ma/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / controversy precedence analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Acontroversy_precedence_analysis.md) | 6.67% | SE ±4.63 pp | 23 / 46 | $0.52 | 109.55source 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` \| 6.667 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.524274 \| USD \| per_test \| Lower is better \| \| `latency` \| 109.553 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.63 pp sourcePopulation 73 stderr 4.632 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/controversy_precedence_analysis/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / insurance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainsurance.md) | 20% | SE ±20 pp | 1 / 49 | $2.44 | 413.09source 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` \| 2.436289 \| USD \| per_test \| Lower is better \| \| `latency` \| 413.094 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/insurance/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena vision / ocr (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Aocr--config-vision.md) | 1,321.77 | — | 10 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1321.766264262086 \| rating \| reported \| Higher is better \| ### Configuration category ocr config vision Reasoning effort high thinkingEnabled true ### Reported statistics lower 1311.4278728307643 rank 10 upper 1332.1046556934077 variance 27.823371894928318 votes 3653 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/1028/row` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / precedents](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aprecedents.md) | 33.81% | SE ±2.43 pp | 7 / 50 | $2.09 | 185.07source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.813 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.090224 \| USD \| per_test \| Lower is better \| \| `latency` \| 185.067 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.43 pp sourcePopulation 80 stderr 2.431 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/precedents/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [MortgageTax / semantic extraction](https://benchize.com/benchmarks/vals-ai%3Amortgage_tax%3Asemantic_extraction.md) | 57.63% | SE ±1.39 pp | 22 / 31 | $0.01 | 4.04source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 57.631 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.010562 \| USD \| per_test \| Lower is better \| \| `latency` \| 4.041 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.39 pp sourcePopulation 98 stderr 1.393 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:58.500Z Evidence SHA-256 `e5c844561f97d119d66d70269e726ecfb68be79eae39acf14ead3e95422e648c` Source locator `/tasks/semantic_extraction/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [MMLU Pro / health](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Ahealth.md) | 84.72% | SE ±1.26 pp | 5 / 40 | $0.00839 | 4.61source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.719 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.008386 \| USD \| per_test \| Lower is better \| \| `latency` \| 4.605 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.26 pp sourcePopulation 138 stderr 1.258 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/health/google~1gemini-3.7-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) | 85.31% | SE ±2.12 pp | 15 / 49 | $3.06 | 664.55source 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` \| 85.305 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.055555 \| USD \| per_test \| Lower is better \| \| `latency` \| 664.552 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.12 pp sourcePopulation 81 stderr 2.124 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__environmental_esg/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.1 / hard](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Ahard.md) | 64.44% | SE ±2.22 pp | 15 / 43 | $3.03 | 1,460.2source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.444 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.032583 \| USD \| per_test \| Lower is better \| \| `latency` \| 1460.201 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.22 pp sourcePopulation 76 stderr 2.222 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/hard/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [SAGE / calculus](https://benchize.com/benchmarks/vals-ai%3Asage%3Acalculus.md) | 83.64% | SE ±4.44 pp | 2 / 36 | $0.04 | 8.82source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 83.642 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.039773 \| USD \| per_test \| Lower is better \| \| `latency` \| 8.815 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.44 pp sourcePopulation 91 stderr 4.441 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/calculus/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / financial modeling](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Afinancial_modeling.md) | 24.16% | SE ±0.33 pp | 17 / 50 | $1.6 | 210.8source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 24.155 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.602877 \| USD \| per_test \| Lower is better \| \| `latency` \| 210.797 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.33 pp sourcePopulation 80 stderr 0.327 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/financial_modeling/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / non english](https://benchize.com/benchmarks/lmarena%3Atext%3Anon_english.md) | 1,486.56 | — | 9 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1486.560534558977 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1480.8249872060828 rank 9 upper 1492.2960819118712 variance 8.563544469115925 votes 17501 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/9739/row` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass forms filings](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__forms_filings.md) | 48.75% | SE ±8.56 pp | 35 / 46 | $0.59 | 125.77source 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` \| 48.751 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.587181 \| USD \| per_test \| Lower is better \| \| `latency` \| 125.77 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.56 pp sourcePopulation 73 stderr 8.555 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__forms_filings/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainternational_trade_sanctions.md) | 0% | SE ±0 pp | 4 / 49 | $2.32 | 912.21source 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` \| 2.317735 \| USD \| per_test \| Lower is better \| \| `latency` \| 912.208 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/international_trade_sanctions/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / english](https://benchize.com/benchmarks/lmarena%3Atext%3Aenglish.md) | 1,483.25 | — | 17 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1483.2512684770538 \| rating \| reported \| Higher is better \| ### Configuration category english config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1476.8194371438694 rank 17 upper 1489.6830998102382 variance 10.768943838647635 votes 11671 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1653/row` |  |  |  |  |  |  |  |
| [LegalBench / interpretation tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Ainterpretation_tasks.md) | 84.61% | SE ±0.35 pp | 5 / 44 | $0.0043 | 2.15source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.608 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.004299 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.145 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.35 pp sourcePopulation 151 stderr 0.348 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/interpretation_tasks/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [TaxEval v2 / overall](https://benchize.com/benchmarks/vals-ai%3Atax_eval_v2%3Aoverall.md) | 74.73% | SE ±0.85 pp | 15 / 40 | $0.02 | 7.83source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_tax_eval_v2.B3yoffr8.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.734 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.018681 \| USD \| per_test \| Lower is better \| \| `latency` \| 7.832 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.85 pp sourcePopulation 145 stderr 0.85 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_eval_v2.B3yoffr8.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:59.000Z Evidence SHA-256 `8b354a6ae848f5bb25a2e8b061b5f576d5b9e0c11b941dc1e6bed078e6529cd6` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [SkillsBench / overall](https://benchize.com/benchmarks/vals-ai%3Askillsbench%3Aoverall.md) | 65.89% | SE ±4.55 pp | 2 / 20 | $1.8 | 185.65source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.891 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.799454 \| USD \| per_test \| Lower is better \| \| `latency` \| 185.651 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.55 pp sourcePopulation 35 stderr 4.554 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:11.147Z Evidence SHA-256 `aaad242949d81b7c7c61c36d4895d4460e9146760c1e6b4b38d235a16043bc61` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / banking finance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Abanking_finance.md) | 20% | SE ±20 pp | 1 / 49 | $2.08 | 435.29source 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` \| 2.07755 \| USD \| per_test \| Lower is better \| \| `latency` \| 435.288 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/banking_finance/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text style control / creative writing](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acreative_writing.md) | 1,490.82 | — | 5 / 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` \| 1490.81605503112 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1482.2433075498147 rank 5 upper 1499.3888025124256 variance 19.13127351055392 votes 6466 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1226/row` |  |  |  |  |  |  |  |
| [MortgageTax / numerical extraction](https://benchize.com/benchmarks/vals-ai%3Amortgage_tax%3Anumerical_extraction.md) | 75.68% | SE ±1.21 pp | 16 / 31 | $0.01 | 4.19source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 75.676 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.01158 \| USD \| per_test \| Lower is better \| \| `latency` \| 4.188 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.21 pp sourcePopulation 98 stderr 1.21 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:58.500Z Evidence SHA-256 `e5c844561f97d119d66d70269e726ecfb68be79eae39acf14ead3e95422e648c` Source locator `/tasks/numerical_extraction/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,500.99 | — | 19 / 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` \| 1500.98704484637 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1487.2275964096389 rank 20 upper 1514.7464932831015 variance 49.283990827436966 votes 2096 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6332/row` |  |  |  |  |  |  |  |
| [Arena text / korean](https://benchize.com/benchmarks/lmarena%3Atext%3Akorean.md) | 1,480.92 | — | 6 / 124 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1480.9200751421197 \| rating \| reported \| Higher is better \| ### Configuration category korean config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1453.820028650833 rank 6 upper 1508.0201216334067 variance 191.18063061709688 votes 524 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8239/row` |  |  |  |  |  |  |  |
| [SAGE / automata](https://benchize.com/benchmarks/vals-ai%3Asage%3Aautomata.md) | 35.32% | SE ±6.14 pp | 19 / 41 | $0.03 | 12.05source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 35.318 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.028498 \| USD \| per_test \| Lower is better \| \| `latency` \| 12.052 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.14 pp sourcePopulation 91 stderr 6.137 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/automata/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__business_commercial.md) | 89.53% | SE ±2.8 pp | 6 / 50 | $0.77 | 298.92source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.525 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.773586 \| USD \| per_test \| Lower is better \| \| `latency` \| 298.916 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.8 pp sourcePopulation 82 stderr 2.803 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__business_commercial/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [MortgageTax / overall](https://benchize.com/benchmarks/vals-ai%3Amortgage_tax%3Aoverall.md) | 66.65% | SE ±0.92 pp | 20 / 31 | $0.01 | 4.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 66.654 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.011071 \| USD \| per_test \| Lower is better \| \| `latency` \| 4.114 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.92 pp sourcePopulation 98 stderr 0.923 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mortgage_tax.DCd-oO5Y.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:58.500Z Evidence SHA-256 `e5c844561f97d119d66d70269e726ecfb68be79eae39acf14ead3e95422e648c` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [LiveCodeBench / overall](https://benchize.com/benchmarks/vals-ai%3Alcb%3Aoverall.md) | 88.65% | SE ±0.92 pp | 5 / 43 | $0.04 | 14.53source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_lcb.miyRUvVa.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.652 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.040218 \| USD \| per_test \| Lower is better \| \| `latency` \| 14.533 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.92 pp sourcePopulation 143 stderr 0.924 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_lcb.miyRUvVa.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:14.220Z Evidence SHA-256 `1a7c7dabc9d95a0182886b1edc42623987cac2095a0d407a5e334f37acdfa2f5` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass rule source lookup](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__rule_source_lookup.md) | 57.9% | SE ±8.82 pp | 17 / 46 | $0.47 | 116.28source 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` \| 57.902 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.471874 \| USD \| per_test \| Lower is better \| \| `latency` \| 116.279 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.82 pp sourcePopulation 73 stderr 8.815 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__rule_source_lookup/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / security](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Asecurity.md) | 33.33% | SE ±6.67 pp | 15 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.333 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.67 pp sourcePopulation 52 stderr 6.667 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/security/google~1gemini-3.7-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 | $2.18 | 678.65source 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` \| 2.18141 \| USD \| per_test \| Lower is better \| \| `latency` \| 678.652 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/structured_finance_securitization/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / operations](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Aoperations.md) | 14.82% | SE ±7.41 pp | 18 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 14.815 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.41 pp sourcePopulation 52 stderr 7.407 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/operations/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate tax](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__tax.md) | 86.62% | SE ±1.97 pp | 6 / 49 | $2.85 | 391.52source 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` \| 86.622 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.850433 \| USD \| per_test \| Lower is better \| \| `latency` \| 391.522 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.97 pp sourcePopulation 81 stderr 1.972 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__tax/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [LegalBench / issue tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aissue_tasks.md) | 92.15% | SE ±0.51 pp | 3 / 44 | $0.00449 | 2.02source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.146 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.004487 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.017 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.51 pp sourcePopulation 151 stderr 0.508 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/issue_tasks/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / polish](https://benchize.com/benchmarks/lmarena%3Atext%3Apolish.md) | 1,503.28 | — | 5 / 117 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1503.2772009573282 \| rating \| reported \| Higher is better \| ### Configuration category polish config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1475.5878038109506 rank 5 upper 1530.9665981037058 variance 199.58634209471634 votes 453 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10152/row` |  |  |  |  |  |  |  |
| [Arena text / spanish](https://benchize.com/benchmarks/lmarena%3Atext%3Aspanish.md) | 1,494.82 | — | 10 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1494.82486656245 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1474.5437088877406 rank 10 upper 1515.1060242371595 variance 107.07530025067423 votes 928 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10774/row` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / science](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Ascience.md) | 11.91% | SE ±4.76 pp | 26 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 11.905 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.76 pp sourcePopulation 52 stderr 4.762 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/science/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass immigration](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__immigration.md) | 83.77% | SE ±6.44 pp | 7 / 50 | $0.58 | 145.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 83.773 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.576237 \| USD \| per_test \| Lower is better \| \| `latency` \| 145.322 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.44 pp sourcePopulation 82 stderr 6.437 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__immigration/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / litigation dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Alitigation_dispute_resolution.md) | 0% | SE ±0 pp | 6 / 49 | $2.59 | 403.07source 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` \| 2.593296 \| USD \| per_test \| Lower is better \| \| `latency` \| 403.072 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/litigation_dispute_resolution/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / media](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Amedia.md) | 0% | SE ±0 pp | 24 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 52 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/media/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [SAGE / overall](https://benchize.com/benchmarks/vals-ai%3Asage%3Aoverall.md) | 49.23% | SE ±3.38 pp | 14 / 41 | $0.03 | 9.79source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 49.232 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.026336 \| USD \| per_test \| Lower is better \| \| `latency` \| 9.791 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.38 pp sourcePopulation 91 stderr 3.382 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.1 / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Aoverall.md) | 77.53% | SE ±0.65 pp | 11 / 43 | $1.02 | 492.2source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.528 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.022219 \| USD \| per_test \| Lower is better \| \| `latency` \| 492.202 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.65 pp sourcePopulation 76 stderr 0.649 tokenTotals {"cache_read_tokens":167065334,"cache_write_tokens":0,"input_tokens":195502339,"output_tokens":3101625,"reasoning_tokens":1363982} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / mathematical sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Amathematical_sciences.md) | 0% | SE ±0 pp | 19 / 30 | — | — | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 41 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/mathematical_sciences/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text style control / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts.md) | 1,507.24 | — | 19 / 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` \| 1507.2353572850466 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1501.6202981861638 rank 20 upper 1512.8504163839293 variance 8.207530044080737 votes 19036 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3470/row` |  |  |  |  |  |  |  |
| [Arena text / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts.md) | 1,495.4 | — | 16 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1495.3993000574887 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1489.8067407383023 rank 16 upper 1500.9918593766752 variance 8.141885986159526 votes 19036 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3466/row` |  |  |  |  |  |  |  |
| [Arena webdev / webdev-html](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev-html.md) | 1,593.42 | — | 23 / 89 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1593.4245613633789 \| rating \| reported \| Higher is better \| ### Configuration category webdev-html config webdev Reasoning effort high thinkingEnabled true ### Reported statistics lower 1578.019449488932 rank 31 upper 1608.8296732378255 variance 61.77795544384882 votes 1850 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/390/row` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aoverall.md) | 7.14% | SE ±3.1 pp | 15 / 30 | $7.19 | 4,825.14source duration | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 7.143 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.190413 \| USD \| per_test \| Lower is better \| \| `latency` \| 4825.144 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.1 pp sourcePopulation 41 stderr 3.1 tokenTotals {"cache_read_tokens":1115979803,"cache_write_tokens":0,"input_tokens":2402435921,"output_tokens":14393442,"reasoning_tokens":3697777} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/overall/google~1gemini-3.7-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) | 84.85% | SE ±2.08 pp | 18 / 49 | $2.38 | 683.69source 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` \| 84.848 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.379245 \| USD \| per_test \| Lower is better \| \| `latency` \| 683.694 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.08 pp sourcePopulation 81 stderr 2.084 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 google/gemini-3.7-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/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text style control / longer query](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Alonger_query.md) | 1,499 | — | 17 / 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` \| 1498.9971663191836 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1492.520547595202 rank 18 upper 1505.4737850431652 variance 10.919442861097483 votes 13535 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8535/row` |  |  |  |  |  |  |  |
| [MMLU Pro / economics](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Aeconomics.md) | 91.71% | SE ±0.95 pp | 9 / 40 | $0.00818 | 3.14source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.706 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.008178 \| USD \| per_test \| Lower is better \| \| `latency` \| 3.142 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.95 pp sourcePopulation 138 stderr 0.949 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/economics/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena vision / diagram (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Adiagram--config-vision.md) | 1,329.11 | — | 11 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1329.1071422962486 \| rating \| reported \| Higher is better \| ### Configuration category diagram config vision Reasoning effort high thinkingEnabled true ### Reported statistics lower 1312.4079691705633 rank 11 upper 1345.806315421934 variance 72.59283415440123 votes 1364 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/477/row` |  |  |  |  |  |  |  |
| [FrontierMath Tiers 1–3 v2 (private)](https://benchize.com/benchmarks/epoch-ai%3Afrontiermath_tiers_1_3_v2.md) | 71.58% | SE ±2.68 pp | 18 / 49 | — | — | [epoch ai · FrontierMath-Tiers-1-3-v2-Private](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tiers_1_3_v2.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.7157894736842105 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.7157894736842105 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion FrontierMath-Tiers-1-3-v2-Private reasoningBudget Not reported Reasoning effort high sourceModelVersion gemini-3.7-flash_high ### Reported statistics Uncertainty Standard error: ±2.68 pp standardError 0.026764156649364653 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tiers_1_3_v2.csv) Source epoch ai Original model label gemini-3.7-flash Source version FrontierMath-Tiers-1-3-v2-Private Evaluated 2026-08-14T19:59:58.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `6e31ece7503f8f89dffe8af33010ed7d1d6f0aa3a80cff14fa79f9b34a7b258f` Source locator `/rows/72` |  |  |  |  |  |  |  |
| [TaxEval v2 / correctness](https://benchize.com/benchmarks/vals-ai%3Atax_eval_v2%3Acorrectness.md) | 63.61% | SE ±1.38 pp | 14 / 40 | $0.02 | 7.83source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_tax_eval_v2.B3yoffr8.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 63.614 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.018681 \| USD \| per_test \| Lower is better \| \| `latency` \| 7.832 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.38 pp sourcePopulation 145 stderr 1.376 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_eval_v2.B3yoffr8.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:59.000Z Evidence SHA-256 `8b354a6ae848f5bb25a2e8b061b5f576d5b9e0c11b941dc1e6bed078e6529cd6` Source locator `/tasks/correctness/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [MedCode / overall](https://benchize.com/benchmarks/vals-ai%3Amedcode%3Aoverall.md) | 53.39% | SE ±2.12 pp | 9 / 49 | $0.04 | 9.16source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_medcode.DehpEDgG.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 53.39 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.038331 \| USD \| per_test \| Lower is better \| \| `latency` \| 9.164 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.12 pp sourcePopulation 107 stderr 2.12 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medcode.DehpEDgG.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:59.516Z Evidence SHA-256 `988a6ae1e07e33ce031572ed4a804be11c36ee97a1dd75cbe4740fff3b534f8c` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / math](https://benchize.com/benchmarks/lmarena%3Atext%3Amath.md) | 1,510.86 | — | 9 / 149 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1510.8553898518658 \| rating \| reported \| Higher is better \| ### Configuration category math config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1495.3778139720796 rank 9 upper 1526.3329657316517 variance 62.36051622473219 votes 1501 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8921/row` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · mathematical](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Amathematical.md) | 1.96% | SE ±1.94 pp | 8 / 9 | $585.39 | — | [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` \| 1.9607843137254901 \| percent \| reported \| Higher is better \| \| `total_cost` \| 585.3949539000001 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 5702475089 \| tokens \| all_trials \| Lower is better \| ### Configuration agent mini-SWE-agent benchmarkVersion v0-1-eval domain mathematical Reasoning effort high ### Reported statistics Uncertainty Standard error: ±1.94 pp accuracyStandardError 1.9414657705424987 passes 1 rank 15 tasks 51 ### 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 Gemini 3.7 Flash Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/14/metrics/domain_metrics/mathematical` |  |  |  |  |  |  |  |
| [CyberBench v1.1 / poc](https://benchize.com/benchmarks/benchmark%3Acyber%3Apoc.md) | 0% | SE ±0 pp | 25 / 28 | — | 0source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `latency` \| 0 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 47 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1.1 Observed 2026-10-21T16:46:15.238Z Evidence SHA-256 `943f593f929f8971a6b746269a2851b5febc2aad4e2f1066895f715033cdc7f1` Source locator `/tasks/poc/google~1gemini-3.7-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) | 92.13% | SE ±1.46 pp | 15 / 49 | $2.81 | 381.56source 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` \| 92.128 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.814369 \| USD \| per_test \| Lower is better \| \| `latency` \| 381.558 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.46 pp sourcePopulation 81 stderr 1.456 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__real_estate/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text style control / non english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Anon_english.md) | 1,485.69 | — | 9 / 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` \| 1485.6906444185568 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text_style_control Reasoning effort high thinkingEnabled true ### Reported statistics lower 1479.952336770034 rank 9 upper 1491.4289520670793 variance 8.571789053603071 votes 17501 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9739/row` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aoverall.md) | 5.71% | SE ±1.6 pp | 7 / 9 | $1,296.59 | — | [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` \| 5.714285714285714 \| percent \| reported \| Higher is better \| \| `total_cost` \| 1296.5871279 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 12023077591 \| tokens \| all_trials \| Lower is better \| ### Configuration agent mini-SWE-agent benchmarkVersion v0-1-eval domain overall Reasoning effort high ### Reported statistics Uncertainty Standard error: ±1.6 pp accuracyStandardError 1.6017483159468233 passes 12 rank 15 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 Gemini 3.7 Flash Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/14/metrics` |  |  |  |  |  |  |  |
| [Mystery Game Puzzles](https://benchize.com/benchmarks/epoch-ai%3Amystery_game_puzzles.md) | 37% | SE ±4.85 pp | 17 / 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` \| 0.37 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.37 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Mystery Game Puzzles reasoningBudget Not reported Reasoning effort high sourceModelVersion gemini-3.7-flash_high ### Reported statistics Uncertainty Standard error: ±4.85 pp standardError 0.048523658709390974 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=mystery_game_puzzles.csv) Source epoch ai Original model label gemini-3.7-flash Source version Mystery Game Puzzles Evaluated 2026-08-14T21:35:50.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `795b977bc82c460a5e2c1d6dd7b2569ddb21853f3829c4f6fdd16edfaf47bfcf` Source locator `/rows/94` |  |  |  |  |  |  |  |
| [FrontierCode 1.1 extended](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Aextended.md) | 54.74% | — | 68 / 125 | $1.79 | — | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.5474 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.606 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.7948 \| USD \| per_rollout \| Lower is better \| ### Configuration harness chisel Reasoning effort high sourceVersion v1_1 subset extended ### Reported statistics flaggedRate 0 sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label Gemini 3.7 Flash Observed 2026-10-21T16:31:00.878Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/Gemini%203.7%20Flash/high/extended` |  |  |  |  |  |  |  |
| [Arena webdev / webdev](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev.md) | 1,590.78 | — | 22 / 90 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1590.77741388594 \| rating \| reported \| Higher is better \| ### Configuration category webdev config webdev Reasoning effort high thinkingEnabled true ### Reported statistics lower 1583.9999661401566 rank 32 upper 1597.5548616317233 variance 11.957383923882299 votes 12636 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/243/row` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass controversy precedence analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__controversy_precedence_analysis.md) | 46% | SE ±8.6 pp | 32 / 46 | $0.52 | 109.55source 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` \| 45.997 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.524274 \| USD \| per_test \| Lower is better \| \| `latency` \| 109.553 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.6 pp sourcePopulation 73 stderr 8.598 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__controversy_precedence_analysis/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / industry entertainment and sports and media](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_entertainment_and_sports_and_media.md) | 1,476.73 | — | 10 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1476.7275125337578 \| rating \| reported \| Higher is better \| ### Configuration category industry_entertainment_and_sports_and_media config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1468.8516409532915 rank 10 upper 1484.6033841142244 variance 16.147342987992705 votes 7798 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/4702/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / administrative regulatory](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aadministrative_regulatory.md) | 42.42% | SE ±6.13 pp | 30 / 50 | $0.7 | 244.66source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 42.424 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.704178 \| USD \| per_test \| Lower is better \| \| `latency` \| 244.664 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.13 pp sourcePopulation 82 stderr 6.13 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/administrative_regulatory/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,511.99 | — | 17 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1511.9887536827302 \| rating \| reported \| Higher is better \| ### Configuration category expert config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1501.2303985286446 rank 19 upper 1522.7471088368159 variance 30.129753050568517 votes 3337 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2489/row` |  |  |  |  |  |  |  |
| [GPQA Diamond](https://benchize.com/benchmarks/gpqa-diamond.md) | 94.82% | SE ±1.35 pp | 6 / 125 | — | — | [epoch ai · GPQA diamond](https://epoch.ai/data/benchmark_data.zip#file=gpqa_diamond.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.9482323232323232 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.9482323232323232 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion GPQA diamond reasoningBudget Not reported Reasoning effort high sourceModelVersion gemini-3.7-flash_high ### Reported statistics Uncertainty Standard error: ±1.35 pp standardError 0.013456525292246259 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=gpqa_diamond.csv) Source epoch ai Original model label gemini-3.7-flash Source version GPQA diamond Evaluated 2026-08-14T18:29:12.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `4b2dd8376cdd249a2c3065603575208fa726c15eabbb1622689a9abc146ba29a` Source locator `/rows/59` |  |  |  |  |  |  |  |
| [Arena text / japanese](https://benchize.com/benchmarks/lmarena%3Atext%3Ajapanese.md) | 1,511.27 | — | 5 / 120 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1511.2722933431983 \| rating \| reported \| Higher is better \| ### Configuration category japanese config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1482.1341461186962 rank 5 upper 1540.4104405677006 variance 221.0180203163931 votes 476 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7955/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate litigation dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__litigation_dispute_resolution.md) | 88.76% | SE ±1.94 pp | 12 / 49 | $2.59 | 403.07source 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` \| 88.764 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.593296 \| USD \| per_test \| Lower is better \| \| `latency` \| 403.072 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.94 pp sourcePopulation 81 stderr 1.936 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__litigation_dispute_resolution/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / all pass](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aall_pass.md) | 47.45% | SE ±0.2 pp | 10 / 50 | $1.48 | 181.33source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.45 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.477884 \| USD \| per_test \| Lower is better \| \| `latency` \| 181.334 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.2 pp sourcePopulation 80 stderr 0.2 tokenTotals {"cache_read_tokens":391282104,"cache_write_tokens":null,"input_tokens":711083728,"output_tokens":16887053,"reasoning_tokens":10058516} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/all_pass/google~1gemini-3.7-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 | $2.29 | 400.15source 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` \| 2.291932 \| USD \| per_test \| Lower is better \| \| `latency` \| 400.146 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/funds_asset_management/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Aoverall.md) | 12.12% | SE ±3.03 pp | 25 / 36 | $8.71 | 5,026.72source duration | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 12.121 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.705677 \| USD \| per_test \| Lower is better \| \| `latency` \| 5026.717 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.03 pp sourcePopulation 52 stderr 3.03 tokenTotals {"cache_read_tokens":1329123269,"cache_write_tokens":0,"input_tokens":1529079806,"output_tokens":10036181,"reasoning_tokens":4455620} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/overall/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [EMB / lbo](https://benchize.com/benchmarks/vals-ai%3Aemb%3Albo.md) | 59.42% | SE ±4.43 pp | 14 / 47 | $10.47 | 1,015.59source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.422 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 10.468715 \| USD \| per_test \| Lower is better \| \| `latency` \| 1015.586 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.43 pp sourcePopulation 76 stderr 4.427 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/lbo/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Arena text / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts_english.md) | 1,486.78 | — | 22 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1486.7825617799622 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text Reasoning effort high thinkingEnabled true ### Reported statistics lower 1479.1902758919173 rank 25 upper 1494.374847668007 variance 15.005446549440606 votes 7632 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Gemini 3.7 Flash (High) Observed 2026-10-21T16:40:19.533Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3892/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 | $1.93 | 265.5source 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` \| 1.928307 \| USD \| per_test \| Lower is better \| \| `latency` \| 265.502 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported 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 google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/intellectual_property/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [EMB / template](https://benchize.com/benchmarks/vals-ai%3Aemb%3Atemplate.md) | 73.6% | SE ±3.44 pp | 11 / 47 | $6.07 | 577.31source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 73.601 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.069648 \| USD \| per_test \| Lower is better \| \| `latency` \| 577.305 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.44 pp sourcePopulation 76 stderr 3.437 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/template/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |
| [Terminal-Bench 2.1 / easy](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Aeasy.md) | 91.67% | SE ±8.33 pp | 15 / 43 | $22.74 | 10,951.5source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.667 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 22.744373 \| USD \| per_test \| Lower is better \| \| `latency` \| 10951.504 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.33 pp sourcePopulation 76 stderr 8.333 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label google/gemini-3.7-flash Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/easy/google~1gemini-3.7-flash` |  |  |  |  |  |  |  |

[Next page](https://benchize.com/models/gemini-3-7-flash-high.md?cursor=502071ef2c906df2099043f23f380c01373a15d7d1de21c38fb35df01a7a014f)

---
Cite as: Gemini 3.7 FlashHigh — Benchize, https://benchize.com/models/gemini-3-7-flash-high (retrieved 2026-10-21)