# Harvey's Legal Agent Benchmark v1 / criteria pass rate insurance > Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__insurance. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__insurance. Tests an agent's ability to complete legal work using documents, spreadsheets, presentations, and file-system tools Canonical URL: https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__insurance Last updated: Wed, 21 Oct 2026 18:52:20 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. Tests an agent's ability to complete legal work using documents, spreadsheets, presentations, and file-system tools Professional work · Models tested: 49 · Score publishers: Vals AI ## Model results | # | Model | Score | ± | Cost | Latency | | | --- | --- | --- | --- | --- | --- | --- | | 1 | [Muse Spark 1.2](https://benchize.com/models/muse-spark-1-2-xhigh.md)Extra high | 94.14% | SE ±1.42 pp | $1.26 | 1,085.91source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.139 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.256553 \| USD \| per_test \| Lower is better \| \| `latency` \| 1085.91 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 131072 provider Meta reasoning Not reported Reasoning effort xhigh Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.42 pp sourcePopulation 81 stderr 1.424 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 meta/muse_spark_1_2 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/meta~1muse_spark_1_2` | | | | | | | | 2 | [Claude Fable 5](https://benchize.com/models/claude-fable-5-max.md)Max | 93.41% | SE ±1.51 pp | $15.61 | 1,417.02source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.407 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 15.611186 \| USD \| per_test \| Lower is better \| \| `latency` \| 1417.02 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.51 pp sourcePopulation 81 stderr 1.505 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 anthropic/claude-fable-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-fable-5` | | | | | | | | 3 | [Gemini 4 Argon](https://benchize.com/models/gemini-4-argon-high.md)High | 93.04% | SE ±1.54 pp | $6.04 | 1,949.21source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.04 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.035978 \| USD \| per_test \| Lower is better \| \| `latency` \| 1949.21 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 262144 provider Google reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.54 pp sourcePopulation 81 stderr 1.543 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-4-argon Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-4-argon` | | | | | | | | 4 | [Muse Wonder](https://benchize.com/models/muse-wonder-max.md)Max | 92.67% | SE ±1.58 pp | $7.52 | 1,483.63source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.674 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.517888 \| USD \| per_test \| Lower is better \| \| `latency` \| 1483.634 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 131072 provider Meta reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.58 pp sourcePopulation 81 stderr 1.58 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 meta/muse-wonder Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/meta~1muse-wonder` | | | | | | | | 5 | [Muse Spark 1.3](https://benchize.com/models/muse-spark-1-3-xhigh.md)Extra high | 92.31% | SE ±1.62 pp | $2.2 | 967.18source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.308 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.204869 \| USD \| per_test \| Lower is better \| \| `latency` \| 967.18 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 131072 provider Meta reasoning Not reported Reasoning effort xhigh Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.62 pp sourcePopulation 81 stderr 1.616 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 meta/muse_spark_1_3 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/meta~1muse_spark_1_3` | | | | | | | | 6 | [Gemini 3.7 Flash](https://benchize.com/models/gemini-3-7-flash-high.md)High | 91.94% | SE ±1.65 pp | $2.44 | 413.09source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.941 \| 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: ±1.65 pp sourcePopulation 81 stderr 1.65 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__insurance/google~1gemini-3.7-flash` | | | | | | | | 7 | [Claude Opus 5](https://benchize.com/models/claude-opus-5-max.md)Max | 91.58% | SE ±1.68 pp | $22.39 | 3,746.53source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.575 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 22.385777 \| USD \| per_test \| Lower is better \| \| `latency` \| 3746.53 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.68 pp sourcePopulation 81 stderr 1.684 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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-opus-5` | | | | | | | | 7 | [MiMo-V2.6-Pro](https://benchize.com/models/mimo-v2-6-pro.md) | 91.58% | SE ±1.68 pp | $0.17 | 863.3source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.575 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.168232 \| USD \| per_test \| Lower is better \| \| `latency` \| 863.304 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider Xiaomi reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.68 pp sourcePopulation 81 stderr 1.684 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 xiaomi/mimo-v2.6-pro Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/xiaomi~1mimo-v2.6-pro` | | | | | | | | 9 | [Step 5 Preview](https://benchize.com/models/step-5-preview.md) | 91.21% | SE ±1.72 pp | $1.09 | 1,827.91source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.209 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.08578 \| USD \| per_test \| Lower is better \| \| `latency` \| 1827.906 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.72 pp sourcePopulation 81 stderr 1.717 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 stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/stepfun~1step-5-preview` | | | | | | | | 10 | [Gemini 3.8 Flash](https://benchize.com/models/gemini-3-8-flash-high.md)High | 90.48% | SE ±1.78 pp | $3.59 | 1,077.37source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.476 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.594349 \| USD \| per_test \| Lower is better \| \| `latency` \| 1077.368 \| 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: ±1.78 pp sourcePopulation 81 stderr 1.78 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-3.8-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.8-flash` | | | | | | | | 11 | [Claude Opus 5.5](https://benchize.com/models/claude-opus-5-5-max.md)Max | 89.86% | SE ±2.05 pp | $16.38 | 2,860.73source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.862 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 16.375474 \| USD \| per_test \| Lower is better \| \| `latency` \| 2860.728 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.05 pp sourcePopulation 81 stderr 2.054 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 anthropic/claude-opus-5-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-opus-5-5` | | | | | | | | 12 | [Claude Sonnet 5.5](https://benchize.com/models/claude-sonnet-5-5-max.md)Max | 89.74% | SE ±1.84 pp | $15.74 | 3,460.94source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.744 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 15.744184 \| USD \| per_test \| Lower is better \| \| `latency` \| 3460.936 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.84 pp sourcePopulation 81 stderr 1.84 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 anthropic/claude-sonnet-5-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-sonnet-5-5` | | | | | | | | 12 | [DeepSeek V4.1 Flash 1017](https://benchize.com/models/deepseek-v4-1-flash-1017-max.md)Max | 89.74% | SE ±1.84 pp | $0.46 | 1,214.74source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.744 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.462519 \| USD \| per_test \| Lower is better \| \| `latency` \| 1214.736 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.84 pp sourcePopulation 81 stderr 1.84 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash-1017 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/deepseek~1deepseek-v4.1-flash-1017` | | | | | | | | 14 | [MiMo-V2.6-Flash](https://benchize.com/models/mimo-v2-6-flash.md) | 89.38% | SE ±1.87 pp | $0.06 | 721.66source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.377 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.060043 \| USD \| per_test \| Lower is better \| \| `latency` \| 721.662 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider Xiaomi reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.87 pp sourcePopulation 81 stderr 1.868 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 xiaomi/mimo-v2.6-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/xiaomi~1mimo-v2.6-flash` | | | | | | | | 14 | [Claude Fable 5.1](https://benchize.com/models/claude-fable-5-1-max.md)Max | 89.38% | SE ±1.87 pp | $32.11 | 5,962.82source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.377 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 32.108773 \| USD \| per_test \| Lower is better \| \| `latency` \| 5962.82 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.87 pp sourcePopulation 81 stderr 1.868 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 anthropic/claude-fable-5-1 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-fable-5-1` | | | | | | | | 16 | [GPT-6 Astra](https://benchize.com/models/gpt-6-astra-max.md)Max | 88.65% | SE ±1.92 pp | $15.66 | 1,289.65source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.645 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 15.655481 \| USD \| per_test \| Lower is better \| \| `latency` \| 1289.65 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.92 pp sourcePopulation 81 stderr 1.924 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 openai/gpt-6-astra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-6-astra` | | | | | | | | 16 | [DeepSeek V4 Pro 0813](https://benchize.com/models/deepseek-v4-pro-0813-max.md)Max | 88.65% | SE ±1.92 pp | $0.12 | 986.99source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.645 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.11707 \| USD \| per_test \| Lower is better \| \| `latency` \| 986.986 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.92 pp sourcePopulation 81 stderr 1.924 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4-pro-0813 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/deepseek~1deepseek-v4-pro-0813` | | | | | | | | 18 | [Muse Spark 1.1](https://benchize.com/models/muse-spark-1-1-xhigh.md)Extra high | 87.93% | SE ±3.04 pp | $0.59 | 491.63source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.931 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.58862 \| USD \| per_test \| Lower is better \| \| `latency` \| 491.63 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 256000 provider Meta reasoning Not reported Reasoning effort xhigh Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.04 pp sourcePopulation 81 stderr 3.038 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 meta/muse_spark_1_1 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/meta~1muse_spark_1_1` | | | | | | | | 19 | [DeepSeek V4 Flash 0731](https://benchize.com/models/deepseek-v4-flash-0731-high.md)High | 87.55% | SE ±2 pp | $0.04 | 607.28source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.546 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.042314 \| USD \| per_test \| Lower is better \| \| `latency` \| 607.276 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2 pp sourcePopulation 81 stderr 2.002 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4-flash-0731 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/deepseek~1deepseek-v4-flash-0731` | | | | | | | | 19 | [DeepSeek V4.1 Flash](https://benchize.com/models/deepseek-v4-1-flash-max.md)Max | 87.55% | SE ±2 pp | $0.15 | 848.14source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.546 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.151326 \| USD \| per_test \| Lower is better \| \| `latency` \| 848.144 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2 pp sourcePopulation 81 stderr 2.002 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/deepseek~1deepseek-v4.1-flash` | | | | | | | | 21 | [Claude Opus 4.7](https://benchize.com/models/claude-opus-4-7-max.md)Max | 87.18% | SE ±2.03 pp | $9.84 | 1,278.74source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.179 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 9.836148 \| USD \| per_test \| Lower is better \| \| `latency` \| 1278.736 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.03 pp sourcePopulation 81 stderr 2.027 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 anthropic/claude-opus-4-7 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-opus-4-7` | | | | | | | | 21 | [MiniMax M3](https://benchize.com/models/minimax-m3.md) | 87.18% | SE ±2.03 pp | $1.05 | 1,205.07source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.179 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.05308 \| USD \| per_test \| Lower is better \| \| `latency` \| 1205.068 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 512000 provider MiniMax reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.03 pp sourcePopulation 81 stderr 2.027 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 minimax/MiniMax-M3 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/minimax~1MiniMax-M3` | | | | | | | | 21 | [Claude Opus 4.8](https://benchize.com/models/claude-opus-4-8-max.md)Max | 87.18% | SE ±2.03 pp | $9.12 | 1,460.75source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.179 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 9.118767 \| USD \| per_test \| Lower is better \| \| `latency` \| 1460.754 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.03 pp sourcePopulation 81 stderr 2.027 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 anthropic/claude-opus-4-8 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-opus-4-8` | | | | | | | | 24 | [Claude Sonnet 5](https://benchize.com/models/claude-sonnet-5-max.md)Max | 86.81% | SE ±2.05 pp | $7.17 | 2,865.77source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.813 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.165675 \| USD \| per_test \| Lower is better \| \| `latency` \| 2865.77 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.05 pp sourcePopulation 81 stderr 2.052 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 anthropic/claude-sonnet-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-sonnet-5` | | | | | | | | 25 | [Gemini 3.6 Flash](https://benchize.com/models/gemini-3-6-flash-high.md)High | 86.45% | SE ±2.08 pp | $1.95 | 1,209.74source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.447 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.949242 \| USD \| per_test \| Lower is better \| \| `latency` \| 1209.736 \| 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 2.075 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-3.6-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.6-flash` | | | | | | | | 26 | [GPT-6.1 Sol](https://benchize.com/models/gpt-6-1-sol-max.md)Max | 86.08% | SE ±2.1 pp | $2.56 | 2,425.11source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.081 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.559806 \| USD \| per_test \| Lower is better \| \| `latency` \| 2425.114 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.1 pp sourcePopulation 81 stderr 2.099 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 openai/gpt-6.1-sol Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-6.1-sol` | | | | | | | | 27 | [Claude Sonnet 4.6](https://benchize.com/models/claude-sonnet-4-6-max.md)Max | 84.62% | SE ±2.19 pp | $2.76 | 999.67source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.615 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.759668 \| USD \| per_test \| Lower is better \| \| `latency` \| 999.666 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning true Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.19 pp sourcePopulation 81 stderr 2.188 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 anthropic/claude-sonnet-4-6 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-sonnet-4-6` | | | | | | | | 28 | [Claude Haiku 5.5](https://benchize.com/models/claude-haiku-5-5-max.md)Max | 84.51% | SE ±2.49 pp | $2.87 | 2,146.63source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.507 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.868148 \| USD \| per_test \| Lower is better \| \| `latency` \| 2146.625 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.49 pp sourcePopulation 81 stderr 2.485 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 anthropic/claude-haiku-5-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-haiku-5-5` | | | | | | | | 29 | [GPT-5.6 Sol](https://benchize.com/models/gpt-5-6-sol-max.md)Max | 84.25% | SE ±2.21 pp | $8.59 | 1,239.26source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.249 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.5933 \| USD \| per_test \| Lower is better \| \| `latency` \| 1239.256 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.21 pp sourcePopulation 81 stderr 2.209 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 openai/gpt-5.6-sol Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-5.6-sol` | | | | | | | | 30 | [Nemotron 3 Ultra](https://benchize.com/models/nemotron-3-ultra-550b-a55b.md) | 82.78% | SE ±2.29 pp | — | 3,435.6source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.784 \| percent \| reported \| Higher is better \| \| `latency` \| 3435.596 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider Nvidia reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.29 pp sourcePopulation 81 stderr 2.289 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 nvidia/nemotron-3-ultra-550b-a55b Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/nvidia~1nemotron-3-ultra-550b-a55b` | | | | | | | | 30 | [GPT-6 Luna](https://benchize.com/models/gpt-6-luna-max.md)Max | 82.78% | SE ±2.29 pp | $0.16 | 824.59source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.784 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.16401 \| USD \| per_test \| Lower is better \| \| `latency` \| 824.586 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.29 pp sourcePopulation 81 stderr 2.289 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 openai/gpt-6-luna Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-6-luna` | | | | | | | | 30 | [GPT-5.5](https://benchize.com/models/gpt-5-5-xhigh.md)Extra high | 82.78% | SE ±2.29 pp | $2.42 | 631.63source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.784 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.417604 \| USD \| per_test \| Lower is better \| \| `latency` \| 631.634 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort xhigh Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.29 pp sourcePopulation 81 stderr 2.289 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 openai/gpt-5.5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-5.5` | | | | | | | | 33 | [GPT-6 Sol](https://benchize.com/models/gpt-6-sol-max.md)Max | 82.42% | SE ±2.31 pp | $2.11 | 762.12source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.418 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.10689 \| USD \| per_test \| Lower is better \| \| `latency` \| 762.124 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.31 pp sourcePopulation 81 stderr 2.308 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 openai/gpt-6-sol Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-6-sol` | | | | | | | | 34 | [GPT-5.6 Luna](https://benchize.com/models/gpt-5-6-luna-max.md)Max | 80.95% | SE ±2.38 pp | $0.31 | 680.89source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.952 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.307965 \| USD \| per_test \| Lower is better \| \| `latency` \| 680.89 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.38 pp sourcePopulation 81 stderr 2.381 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 openai/gpt-5.6-luna Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-5.6-luna` | | | | | | | | 35 | [DeepSeek V4 Pro](https://benchize.com/models/deepseek-v4-pro-max.md)Max | 80.59% | SE ±2.4 pp | $0.63 | 405.38source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.586 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.628366 \| USD \| per_test \| Lower is better \| \| `latency` \| 405.378 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.4 pp sourcePopulation 81 stderr 2.398 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4-pro Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/deepseek~1deepseek-v4-pro` | | | | | | | | 36 | [GPT-5.6 Terra](https://benchize.com/models/gpt-5-6-terra-max.md)Max | 78.76% | SE ±2.48 pp | $2.27 | 899.01source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 78.755 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.265639 \| USD \| per_test \| Lower is better \| \| `latency` \| 899.01 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.48 pp sourcePopulation 81 stderr 2.48 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 openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/openai~1gpt-5.6-terra` | | | | | | | | 37 | [MiMo-V2.5-Pro](https://benchize.com/models/mimo-v2-5-pro.md) | 75.46% | SE ±2.61 pp | $0.09 | 509.88source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 75.458 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.089394 \| USD \| per_test \| Lower is better \| \| `latency` \| 509.878 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider Xiaomi reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.61 pp sourcePopulation 81 stderr 2.609 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 xiaomi/mimo-v2.5-pro Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/xiaomi~1mimo-v2.5-pro` | | | | | | | | 38 | [MiMo-V2.5](https://benchize.com/models/mimo-v2-5.md) | 75.09% | SE ±2.62 pp | $0.03 | 467.07source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 75.092 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.033672 \| USD \| per_test \| Lower is better \| \| `latency` \| 467.072 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 128000 provider Xiaomi reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.62 pp sourcePopulation 81 stderr 2.622 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 xiaomi/mimo-v2.5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/xiaomi~1mimo-v2.5` | | | | | | | | 39 | [Muse Spark Open](https://benchize.com/models/muse-spark-open-xhigh.md)Extra high | 74.73% | SE ±2.64 pp | $6.89 | 7,735.15source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.725 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.893238 \| USD \| per_test \| Lower is better \| \| `latency` \| 7735.152 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 131072 provider Together AI reasoning Not reported Reasoning effort xhigh Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.64 pp sourcePopulation 81 stderr 2.635 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 meta/muse-spark-open Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/meta~1muse-spark-open` | | | | | | | | 40 | [Gemini 3.5 Flash-Lite](https://benchize.com/models/gemini-3-5-flash-lite-high.md)High | 67.03% | SE ±2.85 pp | $0.37 | 358.86source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.033 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.368749 \| USD \| per_test \| Lower is better \| \| `latency` \| 358.856 \| 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: ±2.85 pp sourcePopulation 81 stderr 2.85 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-3.5-flash-lite Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.5-flash-lite` | | | | | | | | 41 | [Gemini 3.5 Flash](https://benchize.com/models/gemini-3-5-flash-high.md)High | 63.74% | SE ±2.92 pp | $1.77 | 401.88source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 63.736 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.769687 \| USD \| per_test \| Lower is better \| \| `latency` \| 401.878 \| 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: ±2.92 pp sourcePopulation 81 stderr 2.915 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-3.5-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.5-flash` | | | | | | | | 42 | [MiniMax M2.7](https://benchize.com/models/minimax-m2-7.md) | 58.24% | SE ±2.99 pp | $0.44 | 412.4source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.242 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.441831 \| USD \| per_test \| Lower is better \| \| `latency` \| 412.404 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 80000 provider MiniMax reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.99 pp sourcePopulation 81 stderr 2.99 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 minimax/MiniMax-M2.7 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/minimax~1MiniMax-M2.7` | | | | | | | | 42 | [Laguna M 1](https://benchize.com/models/laguna-m-1.md) | 58.24% | SE ±2.99 pp | $0.56 | 473.47source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.242 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.555887 \| USD \| per_test \| Lower is better \| \| `latency` \| 473.472 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 32768 provider Poolside reasoning Not reported Reasoning effort unknown Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.99 pp sourcePopulation 81 stderr 2.99 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 poolside/laguna-m.1 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/poolside~1laguna-m.1` | | | | | | | | 44 | [Gemini 3.1 Pro Preview](https://benchize.com/models/gemini-3-1-pro-preview-high.md)High | 56.78% | SE ±3 pp | $0.67 | 257.76source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 56.777 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.666001 \| USD \| per_test \| Lower is better \| \| `latency` \| 257.758 \| 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 pp sourcePopulation 81 stderr 3.004 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-3.1-pro-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.1-pro-preview` | | | | | | | | 45 | [Mistral Medium 3.5](https://benchize.com/models/mistral-medium-3-5-high.md)High | 54.58% | SE ±3.02 pp | $4.7 | 2,551.16source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 54.579 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.703644 \| USD \| per_test \| Lower is better \| \| `latency` \| 2551.164 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 80000 provider Mistral AI reasoning Not reported Reasoning effort high Temperature 0.7 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.02 pp sourcePopulation 81 stderr 3.019 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 mistralai/mistral-medium-3.5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/mistralai~1mistral-medium-3.5` | | | | | | | | 46 | [Gemini 3 Flash Preview](https://benchize.com/models/gemini-3-flash-preview-high.md)High | 46.15% | SE ±3.02 pp | $0.21 | 137.04source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 46.154 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.213394 \| USD \| per_test \| Lower is better \| \| `latency` \| 137.038 \| 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: ±3.02 pp sourcePopulation 81 stderr 3.023 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-flash-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3-flash-preview` | | | | | | | | 47 | [Laguna Xs 2](https://benchize.com/models/laguna-xs-2.md) | 44.84% | SE ±3.34 pp | $0.29 | 1,296.19source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 44.843 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.286257 \| USD \| per_test \| Lower is better \| \| `latency` \| 1296.19 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 32768 provider Poolside reasoning Not reported Reasoning effort unknown Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.34 pp sourcePopulation 81 stderr 3.338 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 poolside/laguna-xs.2 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/poolside~1laguna-xs.2` | | | | | | | | 48 | [Mercury 2.5](https://benchize.com/models/mercury-2-5-high.md)High | 39.19% | SE ±2.96 pp | $0.18 | 103.32source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 39.194 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.182741 \| USD \| per_test \| Lower is better \| \| `latency` \| 103.32 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 65536 provider Inception reasoning Not reported Reasoning effort high Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.96 pp sourcePopulation 81 stderr 2.96 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 inception/mercury-2.5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/inception~1mercury-2.5` | | | | | | | | 49 | [Gemini 3.1 Flash Lite Preview](https://benchize.com/models/gemini-3-1-flash-lite-preview-high.md)High | 23.81% | SE ±2.58 pp | $0.09 | 73.88source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 23.81 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.087874 \| USD \| per_test \| Lower is better \| \| `latency` \| 73.878 \| 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: ±2.58 pp sourcePopulation 81 stderr 2.583 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-3.1-flash-lite-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/google~1gemini-3.1-flash-lite-preview` | | | | | | | --- Cite as: Harvey's Legal Agent Benchmark v1 / criteria pass rate insurance — Benchize, https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__insurance (retrieved 2026-10-21)