# Harvey's Legal Agent Benchmark v1 / criteria pass rate  corporate ma

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_ma. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_ma.

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__corporate_ma
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 1.1](https://benchize.com/models/muse-spark-1-1-xhigh.md)Extra high | 96.15% | SE ±1.09 pp | $0.85 | 778.88source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 96.154 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.847903 \| USD \| per_test \| Lower is better \| \| `latency` \| 778.878 \| 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: ±1.09 pp sourcePopulation 81 stderr 1.09 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__corporate_ma/meta~1muse_spark_1_1` |  |  |  |  |  |  |
| 1 | [Gemini 4 Argon](https://benchize.com/models/gemini-4-argon-high.md)High | 96.15% | SE ±1.09 pp | $8.13 | 2,234.26source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 96.154 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.132612 \| USD \| per_test \| Lower is better \| \| `latency` \| 2234.264 \| 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: ±1.09 pp sourcePopulation 81 stderr 1.09 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__corporate_ma/google~1gemini-4-argon` |  |  |  |  |  |  |
| 3 | [Muse Spark 1.2](https://benchize.com/models/muse-spark-1-2-xhigh.md)Extra high | 95.83% | SE ±1.13 pp | $1.74 | 1,265.26source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 95.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.735278 \| USD \| per_test \| Lower is better \| \| `latency` \| 1265.256 \| 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: ±1.13 pp sourcePopulation 81 stderr 1.133 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__corporate_ma/meta~1muse_spark_1_2` |  |  |  |  |  |  |
| 4 | [Muse Spark 1.3](https://benchize.com/models/muse-spark-1-3-xhigh.md)Extra high | 94.87% | SE ±1.25 pp | $1.93 | 782.07source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.872 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.932419 \| USD \| per_test \| Lower is better \| \| `latency` \| 782.066 \| 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: ±1.25 pp sourcePopulation 81 stderr 1.251 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__corporate_ma/meta~1muse_spark_1_3` |  |  |  |  |  |  |
| 4 | [MiMo-V2.6-Flash](https://benchize.com/models/mimo-v2-6-flash.md) | 94.87% | SE ±1.25 pp | $0.07 | 1,150.2source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.872 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.07413 \| USD \| per_test \| Lower is better \| \| `latency` \| 1150.196 \| 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: ±1.25 pp sourcePopulation 81 stderr 1.251 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__corporate_ma/xiaomi~1mimo-v2.6-flash` |  |  |  |  |  |  |
| 6 | [DeepSeek V4 Flash 0731](https://benchize.com/models/deepseek-v4-flash-0731-high.md)High | 94.23% | SE ±1.32 pp | $0.05 | 597.15source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.231 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.050437 \| USD \| per_test \| Lower is better \| \| `latency` \| 597.146 \| 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: ±1.32 pp sourcePopulation 81 stderr 1.322 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__corporate_ma/deepseek~1deepseek-v4-flash-0731` |  |  |  |  |  |  |
| 7 | [Claude Opus 5](https://benchize.com/models/claude-opus-5-max.md)Max | 93.91% | SE ±1.36 pp | $26 | 3,391.35source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.91 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 25.998286 \| USD \| per_test \| Lower is better \| \| `latency` \| 3391.348 \| 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: ±1.36 pp sourcePopulation 81 stderr 1.356 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__corporate_ma/anthropic~1claude-opus-5` |  |  |  |  |  |  |
| 8 | [Muse Wonder](https://benchize.com/models/muse-wonder-max.md)Max | 93.59% | SE ±1.39 pp | $7.83 | 1,567.18source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.59 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.82908 \| USD \| per_test \| Lower is better \| \| `latency` \| 1567.178 \| 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: ±1.39 pp sourcePopulation 81 stderr 1.389 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__corporate_ma/meta~1muse-wonder` |  |  |  |  |  |  |
| 8 | [MiMo-V2.6-Pro](https://benchize.com/models/mimo-v2-6-pro.md) | 93.59% | SE ±1.39 pp | $0.18 | 2,461.11source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.59 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.176755 \| USD \| per_test \| Lower is better \| \| `latency` \| 2461.11 \| 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: ±1.39 pp sourcePopulation 81 stderr 1.389 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__corporate_ma/xiaomi~1mimo-v2.6-pro` |  |  |  |  |  |  |
| 10 | [Claude Opus 4.7](https://benchize.com/models/claude-opus-4-7-max.md)Max | 93.27% | SE ±1.42 pp | $10.14 | 1,344.03source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.269 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 10.138066 \| USD \| per_test \| Lower is better \| \| `latency` \| 1344.032 \| 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: ±1.42 pp sourcePopulation 81 stderr 1.421 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__corporate_ma/anthropic~1claude-opus-4-7` |  |  |  |  |  |  |
| 10 | [Step 5 Preview](https://benchize.com/models/step-5-preview.md) | 93.27% | SE ±1.42 pp | $0.78 | 1,454.08source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.269 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.778644 \| USD \| per_test \| Lower is better \| \| `latency` \| 1454.082 \| 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: ±1.42 pp sourcePopulation 81 stderr 1.421 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__corporate_ma/stepfun~1step-5-preview` |  |  |  |  |  |  |
| 12 | [Claude Haiku 5.5](https://benchize.com/models/claude-haiku-5-5-max.md)Max | 92.95% | SE ±1.45 pp | $3.46 | 2,112.59source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.949 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.457385 \| USD \| per_test \| Lower is better \| \| `latency` \| 2112.592 \| 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: ±1.45 pp sourcePopulation 81 stderr 1.452 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__corporate_ma/anthropic~1claude-haiku-5-5` |  |  |  |  |  |  |
| 12 | [GPT-6.1 Sol](https://benchize.com/models/gpt-6-1-sol-max.md)Max | 92.95% | SE ±1.45 pp | $4.42 | 3,062.55source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.949 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.419748 \| USD \| per_test \| Lower is better \| \| `latency` \| 3062.552 \| 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: ±1.45 pp sourcePopulation 81 stderr 1.452 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__corporate_ma/openai~1gpt-6.1-sol` |  |  |  |  |  |  |
| 12 | [DeepSeek V4.1 Flash 1017](https://benchize.com/models/deepseek-v4-1-flash-1017-max.md)Max | 92.95% | SE ±1.45 pp | $0.52 | 1,268.74source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.949 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.520056 \| USD \| per_test \| Lower is better \| \| `latency` \| 1268.744 \| 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: ±1.45 pp sourcePopulation 81 stderr 1.452 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__corporate_ma/deepseek~1deepseek-v4.1-flash-1017` |  |  |  |  |  |  |
| 12 | [Claude Opus 4.8](https://benchize.com/models/claude-opus-4-8-max.md)Max | 92.95% | SE ±1.45 pp | $10.04 | 1,524.99source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.949 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 10.04374 \| USD \| per_test \| Lower is better \| \| `latency` \| 1524.986 \| 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: ±1.45 pp sourcePopulation 81 stderr 1.452 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__corporate_ma/anthropic~1claude-opus-4-8` |  |  |  |  |  |  |
| 16 | [Claude Opus 5.5](https://benchize.com/models/claude-opus-5-5-max.md)Max | 92.31% | SE ±1.51 pp | $24.67 | 3,182.29source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.308 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.674742 \| USD \| per_test \| Lower is better \| \| `latency` \| 3182.286 \| 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 1.511 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__corporate_ma/anthropic~1claude-opus-5-5` |  |  |  |  |  |  |
| 16 | [DeepSeek V4 Pro 0813](https://benchize.com/models/deepseek-v4-pro-0813-max.md)Max | 92.31% | SE ±1.51 pp | $0.14 | 1,088.1source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.308 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.144414 \| USD \| per_test \| Lower is better \| \| `latency` \| 1088.096 \| 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: ±1.51 pp sourcePopulation 81 stderr 1.511 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__corporate_ma/deepseek~1deepseek-v4-pro-0813` |  |  |  |  |  |  |
| 16 | [Gemini 3.7 Flash](https://benchize.com/models/gemini-3-7-flash-high.md)High | 92.31% | SE ±1.51 pp | $2.25 | 305.35source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.308 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.25119 \| USD \| per_test \| Lower is better \| \| `latency` \| 305.354 \| 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: ±1.51 pp sourcePopulation 81 stderr 1.511 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__corporate_ma/google~1gemini-3.7-flash` |  |  |  |  |  |  |
| 19 | [DeepSeek V4.1 Flash](https://benchize.com/models/deepseek-v4-1-flash-max.md)Max | 91.99% | SE ±1.54 pp | $0.17 | 444.37source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.987 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.166084 \| USD \| per_test \| Lower is better \| \| `latency` \| 444.366 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 384000 provider DeepSeek reasoning Not reported Reasoning effort max Temperature 1 topP 1 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.54 pp sourcePopulation 81 stderr 1.539 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label deepseek/deepseek-v4.1-flash Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_ma/deepseek~1deepseek-v4.1-flash` |  |  |  |  |  |  |
| 19 | [Claude Fable 5](https://benchize.com/models/claude-fable-5-max.md)Max | 91.99% | SE ±1.54 pp | $18.41 | 1,471.02source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.987 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 18.410087 \| USD \| per_test \| Lower is better \| \| `latency` \| 1471.018 \| 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: ±1.54 pp sourcePopulation 81 stderr 1.539 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-fable-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_ma/anthropic~1claude-fable-5` |  |  |  |  |  |  |
| 21 | [Nemotron 3 Ultra](https://benchize.com/models/nemotron-3-ultra-550b-a55b.md) | 91.67% | SE ±1.57 pp | — | 2,821.81source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.667 \| percent \| reported \| Higher is better \| \| `latency` \| 2821.808 \| 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: ±1.57 pp sourcePopulation 81 stderr 1.567 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__corporate_ma/nvidia~1nemotron-3-ultra-550b-a55b` |  |  |  |  |  |  |
| 22 | [Gemini 3.8 Flash](https://benchize.com/models/gemini-3-8-flash-high.md)High | 91.35% | SE ±1.59 pp | $3.55 | 799.5source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.346 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.550379 \| USD \| per_test \| Lower is better \| \| `latency` \| 799.498 \| 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: ±1.59 pp sourcePopulation 81 stderr 1.594 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__corporate_ma/google~1gemini-3.8-flash` |  |  |  |  |  |  |
| 23 | [Claude Sonnet 4.6](https://benchize.com/models/claude-sonnet-4-6-max.md)Max | 91.03% | SE ±1.62 pp | $3.05 | 1,143.37source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.026 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.051018 \| USD \| per_test \| Lower is better \| \| `latency` \| 1143.366 \| 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: ±1.62 pp sourcePopulation 81 stderr 1.621 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__corporate_ma/anthropic~1claude-sonnet-4-6` |  |  |  |  |  |  |
| 24 | [GPT-6 Astra](https://benchize.com/models/gpt-6-astra-max.md)Max | 90.71% | SE ±1.65 pp | $28.64 | 1,816.78source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.705 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 28.643676 \| USD \| per_test \| Lower is better \| \| `latency` \| 1816.776 \| 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: ±1.65 pp sourcePopulation 81 stderr 1.646 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__corporate_ma/openai~1gpt-6-astra` |  |  |  |  |  |  |
| 25 | [MiniMax M3](https://benchize.com/models/minimax-m3.md) | 90.06% | SE ±1.7 pp | $1.1 | 1,138.96source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.064 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.100807 \| USD \| per_test \| Lower is better \| \| `latency` \| 1138.958 \| 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: ±1.7 pp sourcePopulation 81 stderr 1.696 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__corporate_ma/minimax~1MiniMax-M3` |  |  |  |  |  |  |
| 26 | [Claude Fable 5.1](https://benchize.com/models/claude-fable-5-1-max.md)Max | 89.74% | SE ±1.72 pp | $38.53 | 3,609.03source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.744 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 38.53423 \| USD \| per_test \| Lower is better \| \| `latency` \| 3609.028 \| 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: ±1.72 pp sourcePopulation 81 stderr 1.72 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__corporate_ma/anthropic~1claude-fable-5-1` |  |  |  |  |  |  |
| 26 | [GPT-6 Sol](https://benchize.com/models/gpt-6-sol-max.md)Max | 89.74% | SE ±1.72 pp | $2.74 | 894.13source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.744 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.742624 \| USD \| per_test \| Lower is better \| \| `latency` \| 894.128 \| 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: ±1.72 pp sourcePopulation 81 stderr 1.72 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__corporate_ma/openai~1gpt-6-sol` |  |  |  |  |  |  |
| 28 | [GPT-6 Luna](https://benchize.com/models/gpt-6-luna-max.md)Max | 89.1% | SE ±1.77 pp | $0.33 | 1,091.8source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.103 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.332689 \| USD \| per_test \| Lower is better \| \| `latency` \| 1091.8 \| 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: ±1.77 pp sourcePopulation 81 stderr 1.767 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__corporate_ma/openai~1gpt-6-luna` |  |  |  |  |  |  |
| 29 | [Claude Sonnet 5](https://benchize.com/models/claude-sonnet-5-max.md)Max | 88.46% | SE ±1.81 pp | $7.32 | 2,809.29source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.462 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.318404 \| USD \| per_test \| Lower is better \| \| `latency` \| 2809.288 \| 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: ±1.81 pp sourcePopulation 81 stderr 1.812 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__corporate_ma/anthropic~1claude-sonnet-5` |  |  |  |  |  |  |
| 29 | [DeepSeek V4 Pro](https://benchize.com/models/deepseek-v4-pro-max.md)Max | 88.46% | SE ±1.81 pp | $0.86 | 451.45source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.462 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.855036 \| USD \| per_test \| Lower is better \| \| `latency` \| 451.45 \| 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: ±1.81 pp sourcePopulation 81 stderr 1.812 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__corporate_ma/deepseek~1deepseek-v4-pro` |  |  |  |  |  |  |
| 31 | [GPT-5.5](https://benchize.com/models/gpt-5-5-xhigh.md)Extra high | 87.5% | SE ±1.88 pp | $3.4 | 803.61source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.5 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.395472 \| USD \| per_test \| Lower is better \| \| `latency` \| 803.612 \| 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: ±1.88 pp sourcePopulation 81 stderr 1.875 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__corporate_ma/openai~1gpt-5.5` |  |  |  |  |  |  |
| 31 | [Gemini 3.6 Flash](https://benchize.com/models/gemini-3-6-flash-high.md)High | 87.5% | SE ±1.88 pp | $2.1 | 832.01source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.5 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.104291 \| USD \| per_test \| Lower is better \| \| `latency` \| 832.006 \| 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: ±1.88 pp sourcePopulation 81 stderr 1.875 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__corporate_ma/google~1gemini-3.6-flash` |  |  |  |  |  |  |
| 33 | [GPT-5.6 Luna](https://benchize.com/models/gpt-5-6-luna-max.md)Max | 87.18% | SE ±1.9 pp | $0.4 | 685.32source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.179 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.396871 \| USD \| per_test \| Lower is better \| \| `latency` \| 685.316 \| 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: ±1.9 pp sourcePopulation 81 stderr 1.896 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__corporate_ma/openai~1gpt-5.6-luna` |  |  |  |  |  |  |
| 34 | [GPT-5.6 Terra](https://benchize.com/models/gpt-5-6-terra-max.md)Max | 86.86% | SE ±1.92 pp | $3.07 | 1,028.18source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.859 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.067955 \| USD \| per_test \| Lower is better \| \| `latency` \| 1028.182 \| 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 1.916 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__corporate_ma/openai~1gpt-5.6-terra` |  |  |  |  |  |  |
| 34 | [MiMo-V2.5](https://benchize.com/models/mimo-v2-5.md) | 86.86% | SE ±1.92 pp | $0.04 | 360.64source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.859 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.035857 \| USD \| per_test \| Lower is better \| \| `latency` \| 360.64 \| 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: ±1.92 pp sourcePopulation 81 stderr 1.916 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__corporate_ma/xiaomi~1mimo-v2.5` |  |  |  |  |  |  |
| 36 | [Gemini 3.5 Flash](https://benchize.com/models/gemini-3-5-flash-high.md)High | 86.54% | SE ±1.94 pp | $1.88 | 421.04source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.538 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.877373 \| USD \| per_test \| Lower is better \| \| `latency` \| 421.036 \| 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 1.935 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__corporate_ma/google~1gemini-3.5-flash` |  |  |  |  |  |  |
| 37 | [GPT-5.6 Sol](https://benchize.com/models/gpt-5-6-sol-max.md)Max | 85.58% | SE ±1.99 pp | $12.61 | 1,261.03source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 85.577 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 12.611383 \| USD \| per_test \| Lower is better \| \| `latency` \| 1261.026 \| 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: ±1.99 pp sourcePopulation 81 stderr 1.992 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__corporate_ma/openai~1gpt-5.6-sol` |  |  |  |  |  |  |
| 38 | [MiMo-V2.5-Pro](https://benchize.com/models/mimo-v2-5-pro.md) | 83.97% | SE ±2.08 pp | $0.1 | 505.85source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 83.974 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.104926 \| USD \| per_test \| Lower is better \| \| `latency` \| 505.846 \| 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: ±2.08 pp sourcePopulation 81 stderr 2.08 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__corporate_ma/xiaomi~1mimo-v2.5-pro` |  |  |  |  |  |  |
| 39 | [MiniMax M2.7](https://benchize.com/models/minimax-m2-7.md) | 82.05% | SE ±2.18 pp | $0.44 | 375.64source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.051 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.438626 \| USD \| per_test \| Lower is better \| \| `latency` \| 375.644 \| 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: ±2.18 pp sourcePopulation 81 stderr 2.176 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__corporate_ma/minimax~1MiniMax-M2.7` |  |  |  |  |  |  |
| 40 | [Gemini 3.5 Flash-Lite](https://benchize.com/models/gemini-3-5-flash-lite-high.md)High | 77.24% | SE ±2.38 pp | $0.33 | 248.88source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.244 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.330445 \| USD \| per_test \| Lower is better \| \| `latency` \| 248.88 \| 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: ±2.38 pp sourcePopulation 81 stderr 2.377 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__corporate_ma/google~1gemini-3.5-flash-lite` |  |  |  |  |  |  |
| 41 | [Claude Sonnet 5.5](https://benchize.com/models/claude-sonnet-5-5-max.md)Max | 75.96% | SE ±2.42 pp | $15.02 | 3,174.42source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 75.962 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 15.021633 \| USD \| per_test \| Lower is better \| \| `latency` \| 3174.424 \| 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: ±2.42 pp sourcePopulation 81 stderr 2.423 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__corporate_ma/anthropic~1claude-sonnet-5-5` |  |  |  |  |  |  |
| 42 | [Muse Spark Open](https://benchize.com/models/muse-spark-open-xhigh.md)Extra high | 67.63% | SE ±2.65 pp | $4.07 | 5,887.68source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.628 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.068228 \| USD \| per_test \| Lower is better \| \| `latency` \| 5887.684 \| 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: ±2.65 pp sourcePopulation 81 stderr 2.653 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__corporate_ma/meta~1muse-spark-open` |  |  |  |  |  |  |
| 43 | [Gemini 3.1 Pro Preview](https://benchize.com/models/gemini-3-1-pro-preview-high.md)High | 59.62% | SE ±2.78 pp | $0.46 | 189.98source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.615 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.464081 \| USD \| per_test \| Lower is better \| \| `latency` \| 189.98 \| 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: ±2.78 pp sourcePopulation 81 stderr 2.782 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__corporate_ma/google~1gemini-3.1-pro-preview` |  |  |  |  |  |  |
| 44 | [Laguna M 1](https://benchize.com/models/laguna-m-1.md) | 58.97% | SE ±2.79 pp | $0.72 | 499.32source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.974 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.719898 \| USD \| per_test \| Lower is better \| \| `latency` \| 499.322 \| 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: ±2.79 pp sourcePopulation 81 stderr 2.789 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__corporate_ma/poolside~1laguna-m.1` |  |  |  |  |  |  |
| 45 | [Mistral Medium 3.5](https://benchize.com/models/mistral-medium-3-5-high.md)High | 54.49% | SE ±2.82 pp | $6.68 | 4,108.97source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 54.487 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.680874 \| USD \| per_test \| Lower is better \| \| `latency` \| 4108.97 \| 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: ±2.82 pp sourcePopulation 81 stderr 2.824 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__corporate_ma/mistralai~1mistral-medium-3.5` |  |  |  |  |  |  |
| 46 | [Mercury 2.5](https://benchize.com/models/mercury-2-5-high.md)High | 51.28% | SE ±2.83 pp | $0.2 | 97.04source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 51.282 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.197043 \| USD \| per_test \| Lower is better \| \| `latency` \| 97.04 \| 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: ±2.83 pp sourcePopulation 81 stderr 2.834 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__corporate_ma/inception~1mercury-2.5` |  |  |  |  |  |  |
| 47 | [Gemini 3 Flash Preview](https://benchize.com/models/gemini-3-flash-preview-high.md)High | 41.99% | SE ±2.8 pp | $0.25 | 134.36source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 41.987 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.246824 \| USD \| per_test \| Lower is better \| \| `latency` \| 134.364 \| 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 81 stderr 2.799 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__corporate_ma/google~1gemini-3-flash-preview` |  |  |  |  |  |  |
| 48 | [Laguna Xs 2](https://benchize.com/models/laguna-xs-2.md) | 38.46% | SE ±2.76 pp | $0.23 | 871.13source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 38.462 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.234252 \| USD \| per_test \| Lower is better \| \| `latency` \| 871.134 \| 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: ±2.76 pp sourcePopulation 81 stderr 2.759 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__corporate_ma/poolside~1laguna-xs.2` |  |  |  |  |  |  |
| 49 | [Gemini 3.1 Flash Lite Preview](https://benchize.com/models/gemini-3-1-flash-lite-preview-high.md)High | 32.37% | SE ±2.65 pp | $0.09 | 68.16source duration |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 32.372 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.091726 \| USD \| per_test \| Lower is better \| \| `latency` \| 68.164 \| 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: ±2.65 pp sourcePopulation 81 stderr 2.653 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__corporate_ma/google~1gemini-3.1-flash-lite-preview` |  |  |  |  |  |  |

---
Cite as: Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate ma — Benchize, https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_ma (retrieved 2026-10-21)