# GPT-5.6 TerraMax > Source: Benchize (https://benchize.com) — https://benchize.com/models/gpt-5-6-terra-max. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gpt-5-6-terra-max. OpenAI Canonical URL: https://benchize.com/models/gpt-5-6-terra-max Last updated: Wed, 21 Oct 2026 18:05:10 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. OpenAI ## Benchmarks | Benchmark | Score | ± | Rank | Cost | Latency | Source | | | --- | --- | --- | --- | --- | --- | --- | --- | | [Harvey's Legal Agent Benchmark v1 / corporate governance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acorporate_governance.md) | 0% | SE ±0 pp | 11 / 49 | $3.71 | 895.09source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.705633 \| USD \| per_test \| Lower is better \| \| `latency` \| 895.086 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/corporate_governance/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate funds asset management](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__funds_asset_management.md) | 74.34% | SE ±2.36 pp | 34 / 49 | $3.65 | 1,298.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.344 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.647174 \| USD \| per_test \| Lower is better \| \| `latency` \| 1298.726 \| 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: ±2.36 pp sourcePopulation 81 stderr 2.362 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__funds_asset_management/openai~1gpt-5.6-terra` | | | | | | | | | [Code Migration / overall](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Aoverall.md) | 47.8% | SE ±4.28 pp | 12 / 47 | $8.13 | 2,914.13source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.804 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.127254 \| USD \| per_test \| Lower is better \| \| `latency` \| 2914.132 \| 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: ±4.28 pp sourcePopulation 79 stderr 4.28 tokenTotals {"cache_read_tokens":3610776154,"cache_write_tokens":null,"input_tokens":3758509701,"output_tokens":17992895,"reasoning_tokens":5555226} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate ma](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_ma.md) | 86.86% | SE ±1.92 pp | 34 / 49 | $3.07 | 1,028.18source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.859 \| 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` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / corporate ma](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acorporate_ma.md) | 0% | SE ±0 pp | 21 / 49 | $3.07 | 1,028.18source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.067955 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/corporate_ma/openai~1gpt-5.6-terra` | | | | | | | | | [Chess Puzzles](https://benchize.com/benchmarks/epoch-ai%3Achess_puzzles.md) | 47% | SE ±5.02 pp | 18 / 63 | — | — | [epoch ai · Chess Puzzles](https://epoch.ai/data/benchmark_data.zip#file=chess_puzzles.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.47 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.47 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Chess Puzzles reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±5.02 pp standardError 0.05016135580465918 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=chess_puzzles.csv) Source epoch ai Original model label gpt-5.6-terra Source version Chess Puzzles Evaluated 2026-10-13T13:53:13.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `fa0a449c370b0cceaf273d9472078cce3e5aef6be7009bddf115e6df7f930c54` Source locator `/rows/73` | | | | | | | | | [SAGE / linear algebra](https://benchize.com/benchmarks/vals-ai%3Asage%3Alinear_algebra.md) | 64.03% | SE ±5.38 pp | 22 / 41 | $0.03 | 25.33source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.025 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.033758 \| USD \| per_test \| Lower is better \| \| `latency` \| 25.326 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.38 pp sourcePopulation 91 stderr 5.379 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/linear_algebra/openai~1gpt-5.6-terra` | | | | | | | | | [OTIS Mock AIME 2024–2025](https://benchize.com/benchmarks/epoch-ai%3Aotis_mock_aime_2024_2025.md) | 99.72% | SE ±0.28 pp | 11 / 123 | — | — | [epoch ai · OTIS Mock AIME 2024-2025](https://epoch.ai/data/benchmark_data.zip#file=otis_mock_aime_2024_2025.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.9972222222222222 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.9972222222222222 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion OTIS Mock AIME 2024-2025 reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±0.28 pp standardError 0.002777777777777778 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=otis_mock_aime_2024_2025.csv) Source epoch ai Original model label gpt-5.6-terra Source version OTIS Mock AIME 2024-2025 Evaluated 2026-07-09T17:07:54.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `5b411612f91e155dff4b09d09b548b42f7936482ea8b3c00b510bd46e2f6bc2c` Source locator `/rows/163` | | | | | | | | | [SWE-bench / 1-4 hours](https://benchize.com/benchmarks/vals-ai%3Aswebench%3A1-4%20hours.md) | 97.62% | SE ±2.38 pp | 2 / 39 | $0.67 | 281.06source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 97.619 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.666663 \| USD \| per_test \| Lower is better \| \| `latency` \| 281.058 \| 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 88 stderr 2.381 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:13.711Z Evidence SHA-256 `7a4e2e8485283caac500127419f68ec4bd5414c837ade114fc6afa5b2d34ccaa` Source locator `/tasks/1-4 hours/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate.md) | 79.75% | SE ±0.48 pp | 33 / 49 | $3.2 | 1,024.89source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.751 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.202399 \| USD \| per_test \| Lower is better \| \| `latency` \| 1024.886 \| 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: ±0.48 pp sourcePopulation 81 stderr 0.483 tokenTotals {"cache_read_tokens":740704262,"cache_write_tokens":null,"input_tokens":802970365,"output_tokens":9301238,"reasoning_tokens":2973597} ### 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/openai~1gpt-5.6-terra` | | | | | | | | | [FrontierMath Tier 4 v2 (private)](https://benchize.com/benchmarks/epoch-ai%3Afrontiermath_tier_4_v2.md) | 70.73% | SE ±7.19 pp | 19 / 38 | — | — | [epoch ai · FrontierMath-Tier-4-v2-Private](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tier_4_v2.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.7073170731707317 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.7073170731707317 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion FrontierMath-Tier-4-v2-Private reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±7.19 pp standardError 0.07194088392074452 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tier_4_v2.csv) Source epoch ai Original model label gpt-5.6-terra Source version FrontierMath-Tier-4-v2-Private Evaluated 2026-07-09T15:58:30.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `48b886c828092231989fba6e3819efb7462e89ef681b7b306b4bfa3505829e4a` Source locator `/rows/39` | | | | | | | | | [ProgramBench / partial](https://benchize.com/benchmarks/vals-ai%3Aprogrambench%3Apartial.md) | 72.35% | SE ±1.92 pp | 12 / 40 | $5.38 | 2,016.05source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.345 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.377882 \| USD \| per_test \| Lower is better \| \| `latency` \| 2016.05 \| 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 64 stderr 1.92 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:10.137Z Evidence SHA-256 `e9bbb4378c7a058441a21d456baf9316b96c2cf5b58d371a3c7c803f5000bd8c` Source locator `/tasks/partial/openai~1gpt-5.6-terra` | | | | | | | | | [EMB / lbo](https://benchize.com/benchmarks/vals-ai%3Aemb%3Albo.md) | 50.43% | SE ±6.18 pp | 29 / 47 | $2.79 | 924.29source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 50.428 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.789723 \| USD \| per_test \| Lower is better \| \| `latency` \| 924.286 \| 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: ±6.18 pp sourcePopulation 76 stderr 6.178 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/lbo/openai~1gpt-5.6-terra` | | | | | | | | | [GPQA Diamond](https://benchize.com/benchmarks/gpqa-diamond.md) | 93.31% | SE ±1.54 pp | 13 / 125 | — | — | [epoch ai · GPQA diamond](https://epoch.ai/data/benchmark_data.zip#file=gpqa_diamond.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.9330808080808081 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.9330808080808081 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion GPQA diamond reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±1.54 pp standardError 0.015363672147634115 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=gpqa_diamond.csv) Source epoch ai Original model label gpt-5.6-terra Source version GPQA diamond Evaluated 2026-07-09T17:07:54.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `4b2dd8376cdd249a2c3065603575208fa726c15eabbb1622689a9abc146ba29a` Source locator `/rows/155` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate environmental esg](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__environmental_esg.md) | 71.33% | SE ±2.71 pp | 39 / 49 | $3.03 | 1,195.64source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 71.326 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.029146 \| USD \| per_test \| Lower is better \| \| `latency` \| 1195.644 \| 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: ±2.71 pp sourcePopulation 81 stderr 2.712 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__environmental_esg/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / banking finance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Abanking_finance.md) | 0% | SE ±0 pp | 5 / 49 | $3.56 | 784.63source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.559689 \| USD \| per_test \| Lower is better \| \| `latency` \| 784.634 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/banking_finance/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / fact pattern analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Afact_pattern_analysis.md) | 37.21% | SE ±7.46 pp | 11 / 46 | $4.59 | 3,662.48source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 37.209 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.593033 \| USD \| per_test \| Lower is better \| \| `latency` \| 3662.48 \| 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: ±7.46 pp sourcePopulation 73 stderr 7.458 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/fact_pattern_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainternational_trade_sanctions.md) | 0% | SE ±0 pp | 4 / 49 | $1.85 | 728.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.85423 \| USD \| per_test \| Lower is better \| \| `latency` \| 728.318 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/international_trade_sanctions/openai~1gpt-5.6-terra` | | | | | | | | | [ProgramBench / overall](https://benchize.com/benchmarks/vals-ai%3Aprogrambench%3Aoverall.md) | 0.5% | SE ±0.5 pp | 14 / 40 | $5.38 | 2,016.05source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0.5 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.377882 \| USD \| per_test \| Lower is better \| \| `latency` \| 2016.05 \| 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: ±0.5 pp sourcePopulation 64 stderr 0.5 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_programbench.C_J7ZU4X.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:10.137Z Evidence SHA-256 `e9bbb4378c7a058441a21d456baf9316b96c2cf5b58d371a3c7c803f5000bd8c` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / insurance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainsurance.md) | 0% | SE ±0 pp | 2 / 49 | $2.27 | 899.01source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.265639 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/insurance/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / overall](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aoverall.md) | 30.04% | SE ±3.31 pp | 16 / 46 | $4.83 | 3,630.72source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 30.035 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.829225 \| USD \| per_test \| Lower is better \| \| `latency` \| 3630.72 \| 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: ±3.31 pp sourcePopulation 73 stderr 3.307 tokenTotals {"cache_read_tokens":508089082,"cache_write_tokens":null,"input_tokens":608154133,"output_tokens":46971256,"reasoning_tokens":41166217} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / overall](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aoverall.md) | 41.35% | SE ±3.42 pp | 15 / 50 | $7.91 | 4,172.87source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 41.346 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.907299 \| USD \| per_test \| Lower is better \| \| `latency` \| 4172.869 \| 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: ±3.42 pp sourcePopulation 82 stderr 3.423 tokenTotals {"cache_read_tokens":1375861858,"cache_write_tokens":0,"input_tokens":1570540015,"output_tokens":81682466,"reasoning_tokens":73169720} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [SkillsBench / no skills](https://benchize.com/benchmarks/vals-ai%3Askillsbench%3Ano_skills.md) | 48.84% | SE ±4.57 pp | 3 / 20 | $1.93 | 448.41source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 48.842 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.932564 \| USD \| per_test \| Lower is better \| \| `latency` \| 448.406 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.57 pp sourcePopulation 35 stderr 4.572 tokenTotals {"cache_read_tokens":191843707,"cache_write_tokens":null,"input_tokens":216541737,"output_tokens":3665571,"reasoning_tokens":2365698} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:11.147Z Evidence SHA-256 `aaad242949d81b7c7c61c36d4895d4460e9146760c1e6b4b38d235a16043bc61` Source locator `/tasks/no_skills/openai~1gpt-5.6-terra` | | | | | | | | | [Code Migration / cobol](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Acobol.md) | 70% | SE ±15.28 pp | 1 / 47 | $2.23 | 1,274.91source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 70 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.227453 \| USD \| per_test \| Lower is better \| \| `latency` \| 1274.912 \| 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: ±15.28 pp sourcePopulation 79 stderr 15.275 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/cobol/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 2.1 / easy](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Aeasy.md) | 100% | SE ±0 pp | 1 / 43 | $0.25 | 11,564.19source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 100 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.248207 \| USD \| per_test \| Lower is better \| \| `latency` \| 11564.187 \| 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: ±0 pp sourcePopulation 76 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/easy/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / environmental esg](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aenvironmental_esg.md) | 0% | SE ±0 pp | 1 / 49 | $3.03 | 1,195.64source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.029146 \| USD \| per_test \| Lower is better \| \| `latency` \| 1195.644 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/environmental_esg/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate tax](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__tax.md) | 70.9% | SE ±2.63 pp | 35 / 49 | $3.8 | 1,392.03source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 70.903 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.796478 \| USD \| per_test \| Lower is better \| \| `latency` \| 1392.034 \| 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: ±2.63 pp sourcePopulation 81 stderr 2.631 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__tax/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / weighted pass administrative regulatory](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__administrative_regulatory.md) | 88.76% | SE ±2.29 pp | 10 / 50 | $6.87 | 3,607.9source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 88.759 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.866314 \| USD \| per_test \| Lower is better \| \| `latency` \| 3607.895 \| 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 82 stderr 2.29 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__administrative_regulatory/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / weighted pass criminal](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__criminal.md) | 82.56% | SE ±2.55 pp | 24 / 50 | $8.91 | 4,543.69source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.561 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.912211 \| USD \| per_test \| Lower is better \| \| `latency` \| 4543.691 \| 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: ±2.55 pp sourcePopulation 82 stderr 2.55 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__criminal/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench-Science / mathematical sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Amathematical_sciences.md) | 11.77% | SE ±8.06 pp | 10 / 30 | — | — | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 11.765 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.06 pp sourcePopulation 41 stderr 8.055 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/mathematical_sciences/openai~1gpt-5.6-terra` | | | | | | | | | [SRE Bench / overall](https://benchize.com/benchmarks/vals-ai%3Asrebench%3Aoverall.md) | 9.54% | SE ±1.82 pp | 11 / 24 | $30.86 | 11,800.8source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 9.542 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 30.855656 \| USD \| per_test \| Lower is better \| \| `latency` \| 11800.798 \| 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: ±1.82 pp sourcePopulation 35 stderr 1.819 tokenTotals {"cache_read_tokens":19119371594,"cache_write_tokens":156882339,"input_tokens":19277033725,"output_tokens":38557051,"reasoning_tokens":21920260} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:15.724Z Evidence SHA-256 `a10b2c964f5878b0fc21c888703f311a5ef7d7a74fc3c529cab3ddd9fa474e2b` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Mystery Game Puzzles](https://benchize.com/benchmarks/epoch-ai%3Amystery_game_puzzles.md) | 35% | SE ±4.79 pp | 19 / 71 | — | — | [epoch ai · Mystery Game Puzzles](https://epoch.ai/data/benchmark_data.zip#file=mystery_game_puzzles.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.35 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.35 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Mystery Game Puzzles reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±4.79 pp standardError 0.047937248544110224 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=mystery_game_puzzles.csv) Source epoch ai Original model label gpt-5.6-terra Source version Mystery Game Puzzles Evaluated 2026-07-28T20:58:46.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `795b977bc82c460a5e2c1d6dd7b2569ddb21853f3829c4f6fdd16edfaf47bfcf` Source locator `/rows/123` | | | | | | | | | [EMB / operating](https://benchize.com/benchmarks/vals-ai%3Aemb%3Aoperating.md) | 68.27% | SE ±4.1 pp | 13 / 47 | $1.51 | 683.42source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 68.269 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.509413 \| USD \| per_test \| Lower is better \| \| `latency` \| 683.423 \| 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: ±4.1 pp sourcePopulation 76 stderr 4.095 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/operating/openai~1gpt-5.6-terra` | | | | | | | | | [ARC-AGI-1](https://benchize.com/benchmarks/arc-prize%3Av1_Semi_Private.md) | 96.5% | — | 23 / 213 | 0.55usd per task | — | [arc prize · v1](https://arcprize.org/media/data/leaderboard/v1.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.965 \| score_fraction \| v1_Semi_Private \| Higher is better \| \| `cost_per_task` \| 0.55 \| usd_per_task \| v1_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v1_Semi_Private harness Not reported modelGroup openai-gpt-5-6-terra modelType CoT Reasoning effort max sourceModelVersion openai-gpt-5-6-terra-max thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v1.json) Source arc prize Original model label GPT-5.6 Terra (Max) Source version v1 Observed 2026-10-21T16:34:20.378Z Evidence SHA-256 `302d8267c0522bf450ec1deeac4d060e2cf2fb444ba0a4269014d9535de0bd1f` Source locator `/evaluations/163` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate governance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_governance.md) | 73.29% | SE ±2.59 pp | 39 / 49 | $3.71 | 895.09source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 73.288 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.705633 \| USD \| per_test \| Lower is better \| \| `latency` \| 895.086 \| 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: ±2.59 pp sourcePopulation 81 stderr 2.594 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 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_governance/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 4.0 / security](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Asecurity.md) | 40% | SE ±0 pp | 13 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 40 \| percent \| reported \| Higher 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: ±0 pp sourcePopulation 52 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/security/openai~1gpt-5.6-terra` | | | | | | | | | [Vibe Code Bench 1-100 / overall](https://benchize.com/benchmarks/vals-ai%3Avcb-1-100%3Aoverall.md) | 14.82% | SE ±3.12 pp | 13 / 16 | $33.33 | 6,536.12source duration | [Vals AI · 1.0](https://www.vals.ai/_astro/benchmark_view_vcb_1_100.ZqWqHxGl.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 14.822 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 33.326696 \| USD \| per_test \| Lower is better \| \| `latency` \| 6536.122 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.12 pp sourcePopulation 25 stderr 3.123 tokenTotals {"cache_read_tokens":2199619998,"cache_write_tokens":255401967,"input_tokens":2455464897,"output_tokens":48918335,"reasoning_tokens":30228890} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_vcb_1_100.ZqWqHxGl.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1.0 Observed 2026-10-21T16:46:09.131Z Evidence SHA-256 `4a3ebdcd9364ec3a21193d00de0bf723384d85187244abff859527fd3cc51891` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Finance Agent (v2) / general qualitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Ageneral_qualitative_analysis.md) | 67.8% | SE ±5.94 pp | 15 / 50 | $4.43 | 1,944.23source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.802 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.427551 \| USD \| per_test \| Lower is better \| \| `latency` \| 1944.23 \| 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: ±5.94 pp sourcePopulation 80 stderr 5.94 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/general_qualitative_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 2.1 / medium](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Amedium.md) | 79.39% | SE ±1.6 pp | 17 / 43 | $0.37 | 841.03source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.394 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.367085 \| USD \| per_test \| Lower is better \| \| `latency` \| 841.032 \| 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: ±1.6 pp sourcePopulation 76 stderr 1.603 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/medium/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench Science 0.1 · mathematical](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Amathematical.md) | 9.8% | SE ±4.16 pp | 4 / 9 | $450.66 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 9.803921568627452 \| percent \| reported \| Higher is better \| \| `total_cost` \| 450.6649798000001 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 1639432828 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Codex benchmarkVersion v0-1-eval domain mathematical Reasoning effort max ### Reported statistics Uncertainty Standard error: ±4.16 pp accuracyStandardError 4.163980181949191 passes 5 rank 11 tasks 51 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GPT-5.6 Terra Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/10/metrics/domain_metrics/mathematical` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__international_trade_sanctions.md) | 76.98% | SE ±2.66 pp | 40 / 49 | $1.85 | 728.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 76.984 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.85423 \| USD \| per_test \| Lower is better \| \| `latency` \| 728.318 \| 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: ±2.66 pp sourcePopulation 81 stderr 2.657 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__international_trade_sanctions/openai~1gpt-5.6-terra` | | | | | | | | | [LegalBench / conclusion tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aconclusion_tasks.md) | 90.33% | SE ±0.92 pp | 9 / 44 | $0.00185 | 2.37source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.327 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.001848 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.373 \| 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: ±0.92 pp sourcePopulation 151 stderr 0.916 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/conclusion_tasks/openai~1gpt-5.6-terra` | | | | | | | | | [EMB / ma](https://benchize.com/benchmarks/vals-ai%3Aemb%3Ama.md) | 77.56% | SE ±4.86 pp | 14 / 47 | $1.29 | 762.85source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.558 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.28773 \| USD \| per_test \| Lower is better \| \| `latency` \| 762.847 \| 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: ±4.86 pp sourcePopulation 76 stderr 4.857 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/ma/openai~1gpt-5.6-terra` | | | | | | | | | [DeepSWE](https://benchize.com/benchmarks/deepswe%3Av1.1.md) | 69.62% | Range 67.07–72.18% | 14 / 70 | $4.95 | 1,016.67s | [deepswe](https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_at_1` \| 0.6962305986696231 \| score_fraction \| scored_rollout_attempts \| Higher is better \| \| `pass_at_4` \| 0.8849557522123894 \| score_fraction \| tasks_with_one_passing_rollout \| Higher is better \| \| `cost_per_attempt` \| 4.945847263858093 \| USD \| scored_attempt \| Lower is better \| \| `duration` \| 1016.6698429312638 \| seconds \| scored_attempt \| Lower is better \| ### Configuration harness mini-swe-agent Reasoning effort max sourceConfiguration mini_swe_agent_gpt_5_6_terra_max ### Reported statistics Uncertainty Reported interval: 67.07–72.18% interval [0.6706612091651647,0.7217999881740815] intervalMethod 95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R)) runCount 4 sampleSize 451 taskCount 113 ### Source [Original source](https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json) Source deepswe Original model label gpt-5-6-terra Observed 2026-10-21T16:30:23.371Z Evidence SHA-256 `a7c15d66288fd249c020b9931c017b92d1a3b90e480b3ff34974b752bd030019` Source locator `/rows/13` | | | | | | | | | [FrontierMath Tiers 1–3 v2 (private)](https://benchize.com/benchmarks/epoch-ai%3Afrontiermath_tiers_1_3_v2.md) | 85.96% | SE ±2.06 pp | 10 / 49 | — | — | [epoch ai · FrontierMath-Tiers-1-3-v2-Private](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tiers_1_3_v2.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.8596491228070176 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.8596491228070176 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion FrontierMath-Tiers-1-3-v2-Private reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±2.06 pp standardError 0.020611471958164294 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tiers_1_3_v2.csv) Source epoch ai Original model label gpt-5.6-terra Source version FrontierMath-Tiers-1-3-v2-Private Evaluated 2026-07-09T15:58:30.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `6e31ece7503f8f89dffe8af33010ed7d1d6f0aa3a80cff14fa79f9b34a7b258f` Source locator `/rows/84` | | | | | | | | | [IOI / overall](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aoverall.md) | 87.61% | SE ±6.53 pp | 7 / 29 | $8.66 | 4,992.84source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.611 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.664009 \| USD \| per_test \| Lower is better \| \| `latency` \| 4992.838 \| 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: ±6.53 pp sourcePopulation 44 stderr 6.531 tokenTotals {"cache_read_tokens":8044366,"cache_write_tokens":410842,"input_tokens":8455718,"output_tokens":222620,"reasoning_tokens":182032} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / weighted pass family](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__family.md) | 79.32% | SE ±3.94 pp | 4 / 50 | $9.81 | 5,063.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.323 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 9.812737 \| USD \| per_test \| Lower is better \| \| `latency` \| 5063.727 \| 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: ±3.94 pp sourcePopulation 82 stderr 3.94 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__family/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate energy natural resources](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__energy_natural_resources.md) | 90.76% | SE ±1.67 pp | 30 / 49 | $4.84 | 1,571.69source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.759 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.843052 \| USD \| per_test \| Lower is better \| \| `latency` \| 1571.69 \| 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: ±1.67 pp sourcePopulation 81 stderr 1.666 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__energy_natural_resources/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / administrative regulatory](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aadministrative_regulatory.md) | 57.58% | SE ±6.13 pp | 10 / 50 | $6.87 | 3,607.9source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 57.576 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.866314 \| USD \| per_test \| Lower is better \| \| `latency` \| 3607.895 \| 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: ±6.13 pp sourcePopulation 82 stderr 6.13 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/administrative_regulatory/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate capital markets](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__capital_markets.md) | 86.67% | SE ±2.13 pp | 31 / 49 | $2.62 | 740.83source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.667 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.617447 \| USD \| per_test \| Lower is better \| \| `latency` \| 740.832 \| 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: ±2.13 pp sourcePopulation 81 stderr 2.133 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 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__capital_markets/openai~1gpt-5.6-terra` | | | | | | | | | [SWE-bench / 15 min - 1 hour](https://benchize.com/benchmarks/vals-ai%3Aswebench%3A15%20min%20-%201%20hour.md) | 93.87% | SE ±1.49 pp | 5 / 39 | $0.41 | 183.62source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.87 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.41474 \| USD \| per_test \| Lower is better \| \| `latency` \| 183.618 \| 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: ±1.49 pp sourcePopulation 88 stderr 1.488 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench.Ae8ni7h0.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:13.711Z Evidence SHA-256 `7a4e2e8485283caac500127419f68ec4bd5414c837ade114fc6afa5b2d34ccaa` Source locator `/tasks/15 min - 1 hour/openai~1gpt-5.6-terra` | | | | | | | | | [EMB / template](https://benchize.com/benchmarks/vals-ai%3Aemb%3Atemplate.md) | 68.49% | SE ±4.05 pp | 15 / 47 | $2.26 | 672.66source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 68.49 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.258682 \| USD \| per_test \| Lower is better \| \| `latency` \| 672.659 \| 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: ±4.05 pp sourcePopulation 76 stderr 4.047 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/template/openai~1gpt-5.6-terra` | | | | | | | | | [ProofBench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Aproof_bench%3Aoverall.md) | 74% | SE ±4.41 pp | 14 / 34 | $1.27 | 777.34source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_proof_bench.i3Hwh8rW.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.274209 \| USD \| per_test \| Lower is better \| \| `latency` \| 777.341 \| 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: ±4.41 pp sourcePopulation 52 stderr 4.408 tokenTotals {"cache_read_tokens":222728984,"cache_write_tokens":null,"input_tokens":241131910,"output_tokens":1542468,"reasoning_tokens":704967} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_proof_bench.i3Hwh8rW.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1.1 Observed 2026-10-21T16:46:01.401Z Evidence SHA-256 `22be2aca25de5c4a9eb6448a95713e09a7c97126dfc6f4f75c4afd0e563209cb` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [ARC-AGI-2](https://benchize.com/benchmarks/arc-prize%3Av2_Semi_Private.md) | 83.9% | — | 34 / 212 | 1.09usd per task | — | [arc prize · v2](https://arcprize.org/media/data/leaderboard/v2.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.839 \| score_fraction \| v2_Semi_Private \| Higher is better \| \| `cost_per_task` \| 1.09 \| usd_per_task \| v2_Semi_Private \| Lower is better \| ### Configuration evaluationSplit v2_Semi_Private harness Not reported modelGroup openai-gpt-5-6-terra modelType CoT Reasoning effort max sourceModelVersion openai-gpt-5-6-terra-max thinkingEnabled true ### Source [Original source](https://arcprize.org/media/data/leaderboard/v2.json) Source arc prize Original model label GPT-5.6 Terra (Max) Source version v2 Observed 2026-10-21T16:34:20.750Z Evidence SHA-256 `bca68e12cc9a01954ae21b641844c1ecbabe146cb24b87f0f8e9b1db27433c29` Source locator `/evaluations/165` | | | | | | | | | [EMB / dataroom](https://benchize.com/benchmarks/vals-ai%3Aemb%3Adataroom.md) | 82.16% | SE ±5.74 pp | 19 / 47 | $1.67 | 1,061.01source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.161 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.665623 \| USD \| per_test \| Lower is better \| \| `latency` \| 1061.013 \| 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: ±5.74 pp sourcePopulation 76 stderr 5.744 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/dataroom/openai~1gpt-5.6-terra` | | | | | | | | | [Furniture Assembly Benchmark](https://benchize.com/benchmarks/epoch-ai%3Afurniture_assembly.md) | 54.17% | SE ±6.43 pp | 10 / 21 | — | — | [epoch ai · Furniture Assembly](https://epoch.ai/data/benchmark_data.zip#file=furniture_assembly.csv) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.5416666666666666 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.5416666666666666 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Furniture Assembly reasoningBudget Not reported Reasoning effort max sourceModelVersion gpt-5.6-terra_max ### Reported statistics Uncertainty Standard error: ±6.43 pp standardError 0.06432138878020231 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=furniture_assembly.csv) Source epoch ai Original model label gpt-5.6-terra Source version Furniture Assembly Evaluated 2026-09-10T20:22:34.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `d62ab28005a1b7ab0e57b227d04b0cf86c70a5d8a1981addb26d2ac41c116865` Source locator `/rows/30` | | | | | | | | | [Terminal-Bench 4.0 / ml](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Aml.md) | 15.15% | SE ±3.03 pp | 25 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 15.152 \| percent \| reported \| Higher 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: ±3.03 pp sourcePopulation 52 stderr 3.03 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/ml/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / civil litigation](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Acivil_litigation.md) | 29.63% | SE ±8.96 pp | 26 / 50 | $7.56 | 3,932.08source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 29.63 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.563297 \| USD \| per_test \| Lower is better \| \| `latency` \| 3932.081 \| 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: ±8.96 pp sourcePopulation 82 stderr 8.955 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/civil_litigation/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / funds asset management](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Afunds_asset_management.md) | 0% | SE ±0 pp | 11 / 49 | $3.65 | 1,298.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.647174 \| USD \| per_test \| Lower is better \| \| `latency` \| 1298.726 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/funds_asset_management/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / weighted pass](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass.md) | 65.2% | SE ±3.19 pp | 12 / 46 | $4.83 | 3,630.72source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.199 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.829225 \| USD \| per_test \| Lower is better \| \| `latency` \| 3630.72 \| 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: ±3.19 pp sourcePopulation 73 stderr 3.188 tokenTotals {"cache_read_tokens":508089082,"cache_write_tokens":null,"input_tokens":608154133,"output_tokens":46971256,"reasoning_tokens":41166217} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / energy natural resources](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aenergy_natural_resources.md) | 0% | SE ±0 pp | 28 / 49 | $4.84 | 1,571.69source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.843052 \| USD \| per_test \| Lower is better \| \| `latency` \| 1571.69 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/energy_natural_resources/openai~1gpt-5.6-terra` | | | | | | | | | [Public Benefits Bench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Apublic-benefits-bench%3Aoverall.md) | 62.38% | SE ±1.26 pp | 20 / 34 | $1.2 | 1,171.49source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_snap.CUcEPb0v.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 62.382 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.199723 \| USD \| per_test \| Lower is better \| \| `latency` \| 1171.486 \| 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: ±1.26 pp sourcePopulation 50 stderr 1.26 tokenTotals {"cache_read_tokens":594265088,"cache_write_tokens":null,"input_tokens":641043129,"output_tokens":5293935,"reasoning_tokens":2859562} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_snap.CUcEPb0v.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1.1 Observed 2026-10-21T16:46:17.865Z Evidence SHA-256 `6f091831ab607a26d65d2ade18cba7ce345ea3a67297b8c378e8a06ce398360d` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Finance Agent (v2) / general quantitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Ageneral_quantitative_analysis.md) | 72.68% | SE ±5.61 pp | 25 / 50 | $2.93 | 1,089.19source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.675 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.931004 \| USD \| per_test \| Lower is better \| \| `latency` \| 1089.193 \| 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: ±5.61 pp sourcePopulation 80 stderr 5.607 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/general_quantitative_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / rule source lookup](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Arule_source_lookup.md) | 33.33% | SE ±8.75 pp | 28 / 46 | $4.1 | 2,904.93source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.333 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.09629 \| USD \| per_test \| Lower is better \| \| `latency` \| 2904.931 \| 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: ±8.75 pp sourcePopulation 73 stderr 8.754 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/rule_source_lookup/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / weighted pass controversy precedence analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__controversy_precedence_analysis.md) | 64.83% | SE ±7.96 pp | 10 / 46 | $5.44 | 4,094.14source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.443978 \| USD \| per_test \| Lower is better \| \| `latency` \| 4094.142 \| 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: ±7.96 pp sourcePopulation 73 stderr 7.958 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__controversy_precedence_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 2.1 / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-2-1%3Aoverall.md) | 77.53% | SE ±2.25 pp | 11 / 43 | $0.47 | 519.74source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.528 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.472552 \| USD \| per_test \| Lower is better \| \| `latency` \| 519.739 \| 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: ±2.25 pp sourcePopulation 76 stderr 2.247 tokenTotals {"cache_read_tokens":37760047,"cache_write_tokens":0,"input_tokens":42753712,"output_tokens":2043151,"reasoning_tokens":1345891} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_2_1.CdyoObsA.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2.1 Observed 2026-10-21T16:46:13.216Z Evidence SHA-256 `a5306805cdf86495bdcceff6b2c8534a3768b9853f5077895dab975b809b054f` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate banking finance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__banking_finance.md) | 84.39% | SE ±2.22 pp | 35 / 49 | $3.56 | 784.63source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.387 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.559689 \| USD \| per_test \| Lower is better \| \| `latency` \| 784.634 \| 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: ±2.22 pp sourcePopulation 81 stderr 2.217 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__banking_finance/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench Science 0.1 · earth](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aearth.md) | 8.33% | SE ±5.64 pp | 4 / 9 | $302.41 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 8.333333333333332 \| percent \| reported \| Higher is better \| \| `total_cost` \| 302.40976619999986 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 1197812821 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Codex benchmarkVersion v0-1-eval domain earth Reasoning effort max ### Reported statistics Uncertainty Standard error: ±5.64 pp accuracyStandardError 5.64169333655275 passes 2 rank 11 tasks 24 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GPT-5.6 Terra Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/10/metrics/domain_metrics/earth` | | | | | | | | | [τ³-Banking](https://benchize.com/benchmarks/mercor%3Atau3-banking.md) | 38.8% | — | 17 / 24 | — | — | [mercor](https://www.mercor.com/apex/partner-benchmarks/tau-3-banking-leaderboard/) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 38.8 \| percent \| task_completion_pass_at_1 \| Higher is better \| ### Configuration datasetUrl Not reported datasetVersion Not reported harness Not reported Reasoning effort max ### Reported statistics reportedErrors {"pass_1":8.1} sampleSize 291 ### Source [Original source](https://www.mercor.com/apex/partner-benchmarks/tau-3-banking-leaderboard/) Source mercor Original model label GPT-5.6 Terra Observed 2026-10-21T16:33:27.812Z Evidence SHA-256 `78850adeb8ad047fcb1dd65cc04ffaca61c86fc67e7035f9973805b02483ed07` Source locator `/__NEXT_DATA__/props/pageProps/leaderboard/rows/16` | | | | | | | | | [SkillsBench / overall](https://benchize.com/benchmarks/vals-ai%3Askillsbench%3Aoverall.md) | 58.9% | SE ±4.47 pp | 9 / 20 | $1.75 | 415.07source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.895 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.75247 \| USD \| per_test \| Lower is better \| \| `latency` \| 415.068 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.47 pp sourcePopulation 35 stderr 4.472 tokenTotals {"cache_read_tokens":175914584,"cache_write_tokens":null,"input_tokens":198738807,"output_tokens":3397716,"reasoning_tokens":2179320} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_skillsbench.BE2TXB-M.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:11.147Z Evidence SHA-256 `aaad242949d81b7c7c61c36d4895d4460e9146760c1e6b4b38d235a16043bc61` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / weighted pass rule source lookup](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__rule_source_lookup.md) | 53.29% | SE ±8.77 pp | 29 / 46 | $4.1 | 2,904.93source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 53.288 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.09629 \| USD \| per_test \| Lower is better \| \| `latency` \| 2904.931 \| 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: ±8.77 pp sourcePopulation 73 stderr 8.768 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__rule_source_lookup/openai~1gpt-5.6-terra` | | | | | | | | | [Finance Agent (v2) / earnings analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aearnings_analysis.md) | 77.26% | SE ±5.89 pp | 9 / 50 | $1.42 | 760.58source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.263 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.418824 \| USD \| per_test \| Lower is better \| \| `latency` \| 760.584 \| 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: ±5.89 pp sourcePopulation 80 stderr 5.889 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/earnings_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [SAGE / calculus](https://benchize.com/benchmarks/vals-ai%3Asage%3Acalculus.md) | 59.26% | SE ±6.32 pp | 25 / 36 | $0.06 | 31.53source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.259 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.055115 \| USD \| per_test \| Lower is better \| \| `latency` \| 31.529 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.32 pp sourcePopulation 91 stderr 6.315 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/calculus/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate structured finance securitization](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__structured_finance_securitization.md) | 78.1% | SE ±2.33 pp | 33 / 49 | $3.69 | 1,052.35source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 78.095 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.686565 \| USD \| per_test \| Lower is better \| \| `latency` \| 1052.35 \| 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: ±2.33 pp sourcePopulation 81 stderr 2.334 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__structured_finance_securitization/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / capital markets](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acapital_markets.md) | 0% | SE ±0 pp | 16 / 49 | $2.62 | 740.83source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.617447 \| USD \| per_test \| Lower is better \| \| `latency` \| 740.832 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/capital_markets/openai~1gpt-5.6-terra` | | | | | | | | | [Vibe Code Bench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Avibe-code%3Aoverall.md) | 74.59% | SE ±4.2 pp | 27 / 59 | $7.89 | 1,152.43source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_vibe_code_bench.DcEBMbCA.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.594 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.887015 \| USD \| per_test \| Lower is better \| \| `latency` \| 1152.433 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.2 pp sourcePopulation 116 stderr 4.203 tokenTotals {"cache_read_tokens":1138222288,"cache_write_tokens":null,"input_tokens":1192383214,"output_tokens":4865371,"reasoning_tokens":1379403} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_vibe_code_bench.DcEBMbCA.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1.1 Observed 2026-10-21T16:46:08.625Z Evidence SHA-256 `0e8c23f521a0f751c5592952620e197b376cade41c78578461b7aa5740a173e7` Source locator `/tasks/overall/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / arbitration international dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aarbitration_international_dispute_resolution.md) | 0% | SE ±0 pp | 4 / 49 | $3.31 | 1,162.81source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.307162 \| USD \| per_test \| Lower is better \| \| `latency` \| 1162.808 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/arbitration_international_dispute_resolution/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / weighted pass immigration](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__immigration.md) | 80.69% | SE ±6.94 pp | 17 / 50 | $9.4 | 5,146.75source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.694 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 9.402071 \| USD \| per_test \| Lower is better \| \| `latency` \| 5146.748 \| 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: ±6.94 pp sourcePopulation 82 stderr 6.942 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__immigration/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / antitrust competition](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aantitrust_competition.md) | 0% | SE ±0 pp | 7 / 49 | $3.47 | 1,009.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.466604 \| USD \| per_test \| Lower is better \| \| `latency` \| 1009.726 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/antitrust_competition/openai~1gpt-5.6-terra` | | | | | | | | | [FrontierCode 1.1 extended](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Aextended.md) | 55.84% | — | 60 / 125 | $1.57 | — | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.5584 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.6175 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 1.5696 \| USD \| per_rollout \| Lower is better \| ### Configuration harness codex Reasoning effort max sourceVersion v1_1 subset extended ### Reported statistics flaggedRate 0.0007 sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GPT-5.6 Terra Observed 2026-10-21T16:31:00.878Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/GPT-5.6%20Terra/max/extended` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate real estate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__real_estate.md) | 90.09% | SE ±1.62 pp | 25 / 49 | $4.77 | 1,090.51source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.087 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.771601 \| USD \| per_test \| Lower is better \| \| `latency` \| 1090.508 \| 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: ±1.62 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 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__real_estate/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate healthcare life sciences](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__healthcare_life_sciences.md) | 78.88% | SE ±2.35 pp | 36 / 49 | $3.2 | 965.99source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 78.878 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.200117 \| USD \| per_test \| Lower is better \| \| `latency` \| 965.994 \| 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: ±2.35 pp sourcePopulation 81 stderr 2.349 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__healthcare_life_sciences/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / weighted pass forms filings](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__forms_filings.md) | 68.71% | SE ±7.79 pp | 8 / 46 | $6.68 | 4,910.7source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 68.713 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.681535 \| USD \| per_test \| Lower is better \| \| `latency` \| 4910.699 \| 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: ±7.79 pp sourcePopulation 73 stderr 7.789 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__forms_filings/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate antitrust competition](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__antitrust_competition.md) | 77.54% | SE ±2.52 pp | 36 / 49 | $3.47 | 1,009.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.536 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.466604 \| USD \| per_test \| Lower is better \| \| `latency` \| 1009.726 \| 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: ±2.52 pp sourcePopulation 81 stderr 2.517 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__antitrust_competition/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate white collar defense investigations](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__white_collar_defense_investigations.md) | 73.4% | SE ±2.57 pp | 34 / 49 | $2.56 | 1,094.4source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 73.401 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.560117 \| USD \| per_test \| Lower is better \| \| `latency` \| 1094.404 \| 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: ±2.57 pp sourcePopulation 81 stderr 2.568 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__white_collar_defense_investigations/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 4](https://benchize.com/benchmarks/harbor%3Aterminal-bench-4.md) | 21.52% | 95% CI 18.27–24.77% | 12 / 17 | — | — | [harbor](https://www.tbench.ai/) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 21.52 \| percent \| reported \| Higher is better \| \| `pass_at_2` \| 0.3061 \| score_fraction \| reported \| Higher is better \| \| `pass_at_3` \| 0.3636 \| score_fraction \| reported \| Higher is better \| \| `pass_at_4` \| 0.4061 \| score_fraction \| reported \| Higher is better \| \| `pass_at_5` \| 0.4394 \| score_fraction \| reported \| Higher is better \| ### Configuration agent Codex Reasoning effort max sourceSubmission e53da412-5e92-408c-b369-c767924c7c1c ### Reported statistics Uncertainty 95% confidence interval: 18.27–24.77% confidenceHalfWidth95 3.25 sampleSize 330 ### Source [Original source](https://www.tbench.ai/) Source harbor Original model label GPT-5.6 Terra Evaluated 2026-06-26 Observed 2026-10-21T16:33:12.399Z Evidence SHA-256 `0c2d7608e0711b34acae3e757dc2b8bf58bf1f5f04fec84de6cd02598168712c` Source locator `/rows/26` | | | | | | | | | [Legal Research Bench / constitutional civil rights](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aconstitutional_civil_rights.md) | 34.21% | SE ±7.8 pp | 19 / 50 | $6.78 | 3,655.68source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 34.211 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.783177 \| USD \| per_test \| Lower is better \| \| `latency` \| 3655.679 \| 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: ±7.8 pp sourcePopulation 82 stderr 7.799 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/constitutional_civil_rights/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / data privacy cybersecurity](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Adata_privacy_cybersecurity.md) | 0% | SE ±0 pp | 13 / 49 | $3.07 | 1,345.86source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.070707 \| USD \| per_test \| Lower is better \| \| `latency` \| 1345.862 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/data_privacy_cybersecurity/openai~1gpt-5.6-terra` | | | | | | | | | [SAGE / PDEs](https://benchize.com/benchmarks/vals-ai%3Asage%3APDEs.md) | 38.1% | SE ±6.1 pp | 30 / 41 | $0.04 | 44.22source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 38.095 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.042442 \| USD \| per_test \| Lower is better \| \| `latency` \| 44.223 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 30000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.1 pp sourcePopulation 91 stderr 6.103 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/PDEs/openai~1gpt-5.6-terra` | | | | | | | | | [Tax Agent Bench / numbers calculations](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Anumbers_calculations.md) | 40% | SE ±9.1 pp | 19 / 46 | $2.19 | 1,742.08source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 40 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.186392 \| USD \| per_test \| Lower is better \| \| `latency` \| 1742.081 \| 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: ±9.1 pp sourcePopulation 73 stderr 9.097 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/numbers_calculations/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench-Science / physical sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aphysical_sciences.md) | 11.77% | SE ±8.06 pp | 13 / 30 | — | — | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 11.765 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 128000 provider OpenAI reasoning Not reported Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.06 pp sourcePopulation 41 stderr 8.055 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/physical_sciences/openai~1gpt-5.6-terra` | | | | | | | | | [Legal Research Bench / health](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Ahealth.md) | 80% | SE ±7.43 pp | 1 / 50 | $8.13 | 3,850.23source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 8.132992 \| USD \| per_test \| Lower is better \| \| `latency` \| 3850.231 \| 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: ±7.43 pp sourcePopulation 82 stderr 7.428 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/health/openai~1gpt-5.6-terra` | | | | | | | | | [Terminal-Bench 4.0 / operations](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Aoperations.md) | 7.41% | SE ±3.7 pp | 24 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 7.407 \| percent \| reported \| Higher 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: ±3.7 pp sourcePopulation 52 stderr 3.704 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/operations/openai~1gpt-5.6-terra` | | | | | | | | | [Finance Agent (v2) / disclosure analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Adisclosure_analysis.md) | 66.95% | SE ±6.34 pp | 13 / 50 | $4.01 | 1,555.12source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 66.945 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.00508 \| USD \| per_test \| Lower is better \| \| `latency` \| 1555.117 \| 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: ±6.34 pp sourcePopulation 80 stderr 6.337 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/disclosure_analysis/openai~1gpt-5.6-terra` | | | | | | | | | [LegalBench / rule tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Arule_tasks.md) | 80.53% | SE ±1.64 pp | 20 / 44 | $0.00402 | 5.02source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 80.528 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.004017 \| USD \| per_test \| Lower is better \| \| `latency` \| 5.02 \| 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: ±1.64 pp sourcePopulation 151 stderr 1.635 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/rule_tasks/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / healthcare life sciences](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ahealthcare_life_sciences.md) | 0% | SE ±0 pp | 24 / 49 | $3.2 | 965.99source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.200117 \| USD \| per_test \| Lower is better \| \| `latency` \| 965.994 \| 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: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/healthcare_life_sciences/openai~1gpt-5.6-terra` | | | | | | | | | [LegalBench / issue tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aissue_tasks.md) | 89.39% | SE ±0.57 pp | 13 / 44 | $0.00284 | 2.16source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.392 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.002836 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.159 \| 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: ±0.57 pp sourcePopulation 151 stderr 0.567 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-5.6-terra Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/issue_tasks/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate arbitration international dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__arbitration_international_dispute_resolution.md) | 72.76% | SE ±2.67 pp | 36 / 49 | $3.31 | 1,162.81source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.76 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.307162 \| USD \| per_test \| Lower is better \| \| `latency` \| 1162.808 \| 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: ±2.67 pp sourcePopulation 81 stderr 2.67 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__arbitration_international_dispute_resolution/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / criteria pass rate litigation dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__litigation_dispute_resolution.md) | 81.27% | SE ±2.39 pp | 32 / 49 | $3.25 | 1,024.01source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 81.273 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.252605 \| USD \| per_test \| Lower is better \| \| `latency` \| 1024.006 \| 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: ±2.39 pp sourcePopulation 81 stderr 2.392 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__litigation_dispute_resolution/openai~1gpt-5.6-terra` | | | | | | | | | [Harvey's Legal Agent Benchmark v1 / overall](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aoverall.md) | 0.83% | SE ±0.83 pp | 39 / 49 | $3.2 | 1,024.89source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.202399 \| USD \| per_test \| Lower is better \| \| `latency` \| 1024.886 \| 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: ±0.83 pp sourcePopulation 81 stderr 0.833 tokenTotals {"cache_read_tokens":740704262,"cache_write_tokens":null,"input_tokens":802970365,"output_tokens":9301238,"reasoning_tokens":2973597} ### 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/overall/openai~1gpt-5.6-terra` | | | | | | | | [Next page](https://benchize.com/models/gpt-5-6-terra-max.md?cursor=877f1e29b5a7dda590f7828754f2da7e9f6810784bc378552e9fa34e8fcb7339) --- Cite as: GPT-5.6 TerraMax — Benchize, https://benchize.com/models/gpt-5-6-terra-max (retrieved 2026-10-21)