# Claude Opus 5Max

> Source: Benchize (https://benchize.com) — https://benchize.com/models/claude-opus-5-max. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/claude-opus-5-max.

Anthropic

Canonical URL: https://benchize.com/models/claude-opus-5-max
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

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

Anthropic

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [Legal Research Bench / weighted pass business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__business_commercial.md) | 92.25% | SE ±2.57 pp | 2 / 50 | $7.36 | 2,163.79source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.246 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.357666 \| USD \| per_test \| Lower is better \| \| `latency` \| 2163.785 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.57 pp sourcePopulation 82 stderr 2.569 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__business_commercial/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [MMLU Pro / biology](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Abiology.md) | 95.68% | SE ±0.76 pp | 2 / 40 | $0.02 | 7.06source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 95.676 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.022565 \| USD \| per_test \| Lower is better \| \| `latency` \| 7.056 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.76 pp sourcePopulation 138 stderr 0.76 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/biology/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Earthborne Rangers (EBR-bench)](https://benchize.com/benchmarks/epoch-ai%3Aebr_bench.md) | 45.71% | SE ±3.56 pp | 7 / 13 | — | — | [epoch ai · EBR-bench](https://epoch.ai/data/benchmark_data.zip#file=ebr_bench.csv) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.45714285714285713 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.45714285714285713 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion EBR-bench reasoningBudget Not reported Reasoning effort max sourceModelVersion claude-opus-5_max ### Reported statistics Uncertainty Standard error: ±3.56 pp standardError 0.03563483225498992 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=ebr_bench.csv) Source epoch ai Original model label claude-opus-5 Source version EBR-bench Evaluated 2026-09-05T20:22:12.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `15ec96ff8e424623e6a7452956d849312f595f2b5cb4ebe70ec47831f9087b44` Source locator `/rows/7` |  |  |  |  |  |  |  |
| [SAGE / automata](https://benchize.com/benchmarks/vals-ai%3Asage%3Aautomata.md) | 32.54% | SE ±5.99 pp | 32 / 41 | $0.22 | 74.81source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 32.54 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.221359 \| USD \| per_test \| Lower is better \| \| `latency` \| 74.808 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 30000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.99 pp sourcePopulation 91 stderr 5.991 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/automata/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [CUA-bench / held out sandbox](https://benchize.com/benchmarks/vals-ai%3Acua_bench%3Aheld_out_sandbox.md) | 0% | SE ±0 pp | 3 / 8 | $47.88 | 18.03source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_cua_bench.DpgrKYWA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 47.881221 \| USD \| per_test \| Lower is better \| \| `latency` \| 18.03 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness claude-code maxOutputTokens 128000 provider anthropic reasoning true Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 9 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_cua_bench.DpgrKYWA.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:16.228Z Evidence SHA-256 `8c415c63f28cd80496a4ccca27d1062ec98bb55268e58cba105b8cf2d7ee53d3` Source locator `/tasks/held_out_sandbox/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / overall](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aoverall.md) | 27.14% | SE ±5.35 pp | 8 / 30 | $32.54 | 7,908.45source duration | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 27.143 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 32.543386 \| USD \| per_test \| Lower is better \| \| `latency` \| 7908.453 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Mini-SWE-agent maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.35 pp sourcePopulation 41 stderr 5.354 tokenTotals {"cache_read_tokens":3218944868,"cache_write_tokens":32354267,"input_tokens":6502625236,"output_tokens":30457719,"reasoning_tokens":11809096} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / forms filings](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aforms_filings.md) | 33.19% | SE ±8.72 pp | 3 / 46 | $7.08 | 1,353.85source 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` \| 33.185 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.081583 \| USD \| per_test \| Lower is better \| \| `latency` \| 1353.851 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.72 pp sourcePopulation 73 stderr 8.716 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/forms_filings/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / polish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Apolish.md) | 1,500.54 | — | 8 / 117 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1500.5448048923674 \| rating \| reported \| Higher is better \| ### Configuration category polish config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1475.1677182786627 rank 8 upper 1525.9218915060721 variance 167.64374032339708 votes 526 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10155/row` |  |  |  |  |  |  |  |
| [ArXivMath 06/2026](https://benchize.com/benchmarks/matharena%3Aarxiv--june.md) | 82.64% | 95% CI 76.45–88.83% | 10 / 22 | 3.66usd per problem | — | [matharena · arxiv--june](https://matharena.ai/competition_tables/arxiv--june) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.64 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 3.66 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition arxiv--june predicted false Reasoning effort max sourceModelName Claude-Opus-5 (max) sourceModelVersion anthropic/opus_5_max thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 76.45–88.83% confidenceInterval95 6.19 problems 48 ### Source [Original source](https://matharena.ai/competition_tables/arxiv--june) Source matharena Original model label Claude-Opus-5 (max) Source version arxiv--june Observed 2026-10-21T16:52:29.905Z Evidence SHA-256 `6832faff9be7e16a6c646c59017e30a2caedebe8cf02caf617a5ed9d8fb03bab` Source locator `/table/rows/9` |  |  |  |  |  |  |  |
| [Tax Agent Bench / controversy precedence analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Acontroversy_precedence_analysis.md) | 23.3% | SE ±7.84 pp | 4 / 46 | $5.66 | 1,182.24source 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` \| 23.295 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.663294 \| USD \| per_test \| Lower is better \| \| `latency` \| 1182.24 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.84 pp sourcePopulation 73 stderr 7.841 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/controversy_precedence_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [EMB / template](https://benchize.com/benchmarks/vals-ai%3Aemb%3Atemplate.md) | 74.69% | SE ±3.39 pp | 10 / 47 | $6.51 | 1,494.85source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.693 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.50608 \| USD \| per_test \| Lower is better \| \| `latency` \| 1494.85 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.39 pp sourcePopulation 76 stderr 3.393 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/template/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / earth sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aearth_sciences.md) | 12.5% | SE ±12.5 pp | 11 / 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` \| 12.5 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Mini-SWE-agent maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±12.5 pp sourcePopulation 41 stderr 12.5 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 anthropic/claude-opus-5 Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/earth_sciences/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [CUA-bench / held out sports](https://benchize.com/benchmarks/vals-ai%3Acua_bench%3Aheld_out_sports.md) | 0% | SE ±0 pp | 1 / 8 | $26.6 | 42.38source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_cua_bench.DpgrKYWA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 26.600182 \| USD \| per_test \| Lower is better \| \| `latency` \| 42.38 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness claude-code maxOutputTokens 128000 provider anthropic reasoning true Reasoning effort max Temperature Not reported topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 9 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_cua_bench.DpgrKYWA.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:16.228Z Evidence SHA-256 `8c415c63f28cd80496a4ccca27d1062ec98bb55268e58cba105b8cf2d7ee53d3` Source locator `/tasks/held_out_sports/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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 | $28.23 | 3,845.52source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 28.228604 \| USD \| per_test \| Lower is better \| \| `latency` \| 3845.516 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/funds_asset_management/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate insurance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__insurance.md) | 91.58% | SE ±1.68 pp | 7 / 49 | $22.39 | 3,746.53source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.575 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 22.385777 \| USD \| per_test \| Lower is better \| \| `latency` \| 3746.53 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.68 pp sourcePopulation 81 stderr 1.684 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__insurance/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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 | $21.03 | 2,729.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 21.027612 \| USD \| per_test \| Lower is better \| \| `latency` \| 2729.106 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/arbitration_international_dispute_resolution/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena webdev / webdev](https://benchize.com/benchmarks/lmarena%3Awebdev%3Awebdev.md) | 1,688.86 | — | 7 / 90 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1688.8555367369042 \| rating \| reported \| Higher is better \| ### Configuration category webdev config webdev Reasoning effort max thinkingEnabled true ### Reported statistics lower 1682.6517812580007 rank 8 upper 1695.0592922158078 variance 10.01874127471985 votes 18555 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/219/row` |  |  |  |  |  |  |  |
| [Arena text style control / longer query](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Alonger_query.md) | 1,502.5 | — | 12 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1502.4955746621254 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1496.252058354016 rank 13 upper 1508.7390909702347 variance 10.147576144675007 votes 15833 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8530/row` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / general quantitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Ageneral_quantitative_analysis.md) | 78.5% | SE ±0.94 pp | 13 / 50 | $4.85 | 474.84source 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` \| 78.502 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.848634 \| USD \| per_test \| Lower is better \| \| `latency` \| 474.835 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.94 pp sourcePopulation 80 stderr 0.938 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/general_quantitative_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass.md) | 75.06% | SE ±2.96 pp | 3 / 46 | $5.14 | 1,008.31source 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` \| 75.06 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.136554 \| USD \| per_test \| Lower is better \| \| `latency` \| 1008.306 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.96 pp sourcePopulation 73 stderr 2.956 tokenTotals {"cache_read_tokens":250544336,"cache_write_tokens":71504899,"input_tokens":322077193,"output_tokens":16761490,"reasoning_tokens":45865} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Vals Science Index / mysterymechanism](https://benchize.com/benchmarks/vals-ai%3Ascience_index%3Amysterymechanism.md) | 37.39% | SE ±3.26 pp | 6 / 17 | $3.28 | 1,012.54source duration | [Vals AI · 1.0](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 37.387 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.280693 \| USD \| per_test \| Lower is better \| \| `latency` \| 1012.543 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.26 pp sourcePopulation 21 stderr 3.255 tokenTotals {"cache_read_tokens":234441635,"cache_write_tokens":25563677,"input_tokens":260014806,"output_tokens":18050903,"reasoning_tokens":13538433} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1.0 Observed 2026-10-21T16:45:50.879Z Evidence SHA-256 `fa4313008e91752fcfc185aaa78665f780d0fc23d7c7198a9853dbf2eaaaef1f` Source locator `/tasks/mysterymechanism/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [LegalBench / conclusion tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aconclusion_tasks.md) | 91.56% | SE ±0.7 pp | 3 / 44 | $0.00538 | 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` \| 91.56 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.005379 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.373 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.7 pp sourcePopulation 151 stderr 0.696 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/conclusion_tasks/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [LiveCodeBench / hard](https://benchize.com/benchmarks/vals-ai%3Alcb%3Ahard.md) | 77.43% | SE ±2.24 pp | 4 / 43 | $0.35 | 153.77source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_lcb.miyRUvVa.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.429 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.348014 \| USD \| per_test \| Lower is better \| \| `latency` \| 153.772 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.24 pp sourcePopulation 143 stderr 2.235 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_lcb.miyRUvVa.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:14.220Z Evidence SHA-256 `1a7c7dabc9d95a0182886b1edc42623987cac2095a0d407a5e334f37acdfa2f5` Source locator `/tasks/hard/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [MMLU Pro / computer science](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Acomputer_science.md) | 93.9% | SE ±1.18 pp | 3 / 40 | $0.03 | 8.9source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.902 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.028057 \| USD \| per_test \| Lower is better \| \| `latency` \| 8.895 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.18 pp sourcePopulation 138 stderr 1.182 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/computer_science/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / immigration](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aimmigration.md) | 0% | SE ±0 pp | 6 / 49 | $19.86 | 2,995.12source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 19.864027 \| USD \| per_test \| Lower is better \| \| `latency` \| 2995.124 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/immigration/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [FrontierMath Tiers 1–3 v2 (private)](https://benchize.com/benchmarks/epoch-ai%3Afrontiermath_tiers_1_3_v2.md) | 85.61% | SE ±2.08 pp | 12 / 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` \| 0.856140350877193 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.856140350877193 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion FrontierMath-Tiers-1-3-v2-Private reasoningBudget Not reported Reasoning effort max sourceModelVersion claude-opus-5_max ### Reported statistics Uncertainty Standard error: ±2.08 pp standardError 0.020824894575343943 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=frontiermath_tiers_1_3_v2.csv) Source epoch ai Original model label claude-opus-5 Source version FrontierMath-Tiers-1-3-v2-Private Evaluated 2026-07-24T19:54:20.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `6e31ece7503f8f89dffe8af33010ed7d1d6f0aa3a80cff14fa79f9b34a7b258f` Source locator `/rows/81` |  |  |  |  |  |  |  |
| [Arena text style control / exclude ties](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aexclude_ties.md) | 1,505.7 | — | 14 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1505.703125716067 \| rating \| reported \| Higher is better \| ### Configuration category exclude_ties config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1499.456338918511 rank 14 upper 1511.9499125136228 variance 10.158209970624737 votes 25329 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2067/row` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / hardware](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Ahardware.md) | 33.33% | SE ±6.67 pp | 9 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.333 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.67 pp sourcePopulation 52 stderr 6.667 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/hardware/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / litigation dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Alitigation_dispute_resolution.md) | 0% | SE ±0 pp | 6 / 49 | $19.38 | 2,867.04source 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` \| 19.378797 \| USD \| per_test \| Lower is better \| \| `latency` \| 2867.042 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/litigation_dispute_resolution/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass fact pattern analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__fact_pattern_analysis.md) | 77.34% | SE ±6.16 pp | 2 / 46 | $4.48 | 962.38source 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` \| 77.336 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.479733 \| USD \| per_test \| Lower is better \| \| `latency` \| 962.376 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.16 pp sourcePopulation 73 stderr 6.157 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__fact_pattern_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Furniture Assembly Benchmark](https://benchize.com/benchmarks/epoch-ai%3Afurniture_assembly.md) | 60.83% | SE ±6.19 pp | 7 / 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` \| 0.6083333333333333 \| score_fraction \| Best score (across scorers) \| Higher is better \| \| `mean_score` \| 0.6083333333333333 \| score_fraction \| mean_score \| Higher is better \| ### Configuration benchmarkVersion Furniture Assembly reasoningBudget Not reported Reasoning effort max sourceModelVersion claude-opus-5_max ### Reported statistics Uncertainty Standard error: ±6.19 pp standardError 0.06185876325418331 ### Source [Original source](https://epoch.ai/data/benchmark_data.zip#file=furniture_assembly.csv) Source epoch ai Original model label claude-opus-5 Source version Furniture Assembly Evaluated 2026-09-10T20:23:29.000Z Observed 2026-10-21T16:30:24.199Z Evidence SHA-256 `d62ab28005a1b7ab0e57b227d04b0cf86c70a5d8a1981addb26d2ac41c116865` Source locator `/rows/22` |  |  |  |  |  |  |  |
| [Arena text / chinese](https://benchize.com/benchmarks/lmarena%3Atext%3Achinese.md) | 1,572.96 | — | 4 / 146 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1572.9562190506795 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1560.6026917673544 rank 4 upper 1585.3097463340043 variance 39.727000460798465 votes 2681 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/420/row` |  |  |  |  |  |  |  |
| [Time Horizon Index: KSP / mean progress to date](https://benchize.com/benchmarks/vals-ai%3Atime_horizon_index%3Amean_progress_to_date.md) | 17.53% | SE ±0 pp | 5 / 10 | $1,348.77 | 0source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_ksp_bench.CDDTSam7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 17.527 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1348.765783 \| USD \| per_test \| Lower is better \| \| `latency` \| 0 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 13 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ksp_bench.CDDTSam7.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:16.730Z Evidence SHA-256 `be251b72ecceae7912b9577f9b3a3512740c308a34f81ac823952e919adae50f` Source locator `/tasks/mean_progress_to_date/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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) | 86.25% | SE ±2.1 pp | 29 / 49 | $24.31 | 3,388.15source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.245 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.310288 \| USD \| per_test \| Lower is better \| \| `latency` \| 3388.15 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.1 pp sourcePopulation 81 stderr 2.104 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__banking_finance/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / fact pattern analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Afact_pattern_analysis.md) | 48.84% | SE ±7.71 pp | 3 / 46 | $4.48 | 962.38source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 48.837 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 4.479733 \| USD \| per_test \| Lower is better \| \| `latency` \| 962.376 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.71 pp sourcePopulation 73 stderr 7.713 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/fact_pattern_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / healthcare life sciences](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ahealthcare_life_sciences.md) | 20% | SE ±20 pp | 5 / 49 | $24.92 | 3,583.41source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.921853 \| USD \| per_test \| Lower is better \| \| `latency` \| 3583.414 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/healthcare_life_sciences/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_mathematical.md) | 1,523.71 | — | 6 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1523.7107411988004 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1509.4304031337551 rank 6 upper 1537.9910792638457 variance 53.086096915423894 votes 1899 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/5918/row` |  |  |  |  |  |  |  |
| [IOI / ioi2026](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aioi2026.md) | 65.83% | SE ±14.58 pp | 10 / 29 | $23.57 | 4,315.54source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.568374 \| USD \| per_test \| Lower is better \| \| `latency` \| 4315.537 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±14.58 pp sourcePopulation 44 stderr 14.58 tokenTotals {"cache_read_tokens":137949530,"cache_write_tokens":4373778,"input_tokens":142324076,"output_tokens":1803821,"reasoning_tokens":1472640} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/ioi2026/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / hard prompts english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ahard_prompts_english.md) | 1,512.42 | — | 16 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1512.415313954276 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts_english config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1505.0689035330383 rank 17 upper 1519.761724375514 variance 14.049284034110352 votes 8691 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/3884/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / intellectual property](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aintellectual_property.md) | 0% | SE ±0 pp | 6 / 49 | $23.47 | 3,194.55source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.467391 \| USD \| per_test \| Lower is better \| \| `latency` \| 3194.55 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/intellectual_property/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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) | 72.04% | SE ±2.69 pp | 38 / 49 | $24.42 | 4,134.27source 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` \| 72.043 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.419754 \| USD \| per_test \| Lower is better \| \| `latency` \| 4134.27 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.69 pp sourcePopulation 81 stderr 2.692 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__environmental_esg/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate emerging companies venture capital](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__emerging_companies_venture_capital.md) | 89.08% | SE ±1.83 pp | 16 / 49 | $21.32 | 3,356.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.078 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 21.32291 \| USD \| per_test \| Lower is better \| \| `latency` \| 3356.11 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.83 pp sourcePopulation 81 stderr 1.825 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__emerging_companies_venture_capital/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate arbitration international dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__arbitration_international_dispute_resolution.md) | 90.32% | SE ±1.77 pp | 10 / 49 | $21.03 | 2,729.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.323 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 21.027612 \| USD \| per_test \| Lower is better \| \| `latency` \| 2729.106 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.77 pp sourcePopulation 81 stderr 1.773 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__arbitration_international_dispute_resolution/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / instruction following](https://benchize.com/benchmarks/lmarena%3Atext%3Ainstruction_following.md) | 1,518.02 | — | 5 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1518.0189485152716 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1511.3665892476783 rank 5 upper 1524.6713077828651 variance 11.520072423199881 votes 11699 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7538/row` |  |  |  |  |  |  |  |
| [Code Migration / cobol](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Acobol.md) | 70% | SE ±15.28 pp | 1 / 47 | $17.5 | 5,017.8source 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` \| 17.499254 \| USD \| per_test \| Lower is better \| \| `latency` \| 5017.804 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/cobol/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,552.44 | — | 2 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1552.4387802655021 \| rating \| reported \| Higher is better \| ### Configuration category expert config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1541.7335572997574 rank 2 upper 1563.1440032312466 variance 29.832884874084613 votes 3597 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2472/row` |  |  |  |  |  |  |  |
| [EMB / dataroom](https://benchize.com/benchmarks/vals-ai%3Aemb%3Adataroom.md) | 87.68% | SE ±3.69 pp | 1 / 47 | $5.42 | 1,405.31source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.681 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.420298 \| USD \| per_test \| Lower is better \| \| `latency` \| 1405.306 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.69 pp sourcePopulation 76 stderr 3.687 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/dataroom/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / spanish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aspanish.md) | 1,491.02 | — | 9 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1491.0227786647183 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1471.2249045744043 rank 9 upper 1510.8206527550324 variance 102.03306524709795 votes 993 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10773/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate energy natural resources](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__energy_natural_resources.md) | 95.05% | SE ±1.25 pp | 16 / 49 | $31.65 | 3,747.78source 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` \| 95.05 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 31.649398 \| USD \| per_test \| Lower is better \| \| `latency` \| 3747.78 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.25 pp sourcePopulation 81 stderr 1.248 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__energy_natural_resources/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [BioMysteryBench / overall](https://benchize.com/benchmarks/vals-ai%3Abiomysterybench%3Aoverall.md) | 79.26% | SE ±2.06 pp | 3 / 20 | $3.29 | 2,195.96source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 79.259 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.292774 \| USD \| per_test \| Lower is better \| \| `latency` \| 2195.964 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Terminus 2 maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.06 pp sourcePopulation 27 stderr 2.062 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:04.029Z Evidence SHA-256 `59bf79d38ec415a34ffb1c2901a16ecbbe101b7fa864e86104d2571fc7cfc23d` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [MedCode / overall](https://benchize.com/benchmarks/vals-ai%3Amedcode%3Aoverall.md) | 63.57% | SE ±1.99 pp | 1 / 49 | $0.16 | 24.5source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_medcode.DehpEDgG.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 63.57 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.156845 \| USD \| per_test \| Lower is better \| \| `latency` \| 24.5 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 30000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.99 pp sourcePopulation 107 stderr 1.993 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medcode.DehpEDgG.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:59.516Z Evidence SHA-256 `988a6ae1e07e33ce031572ed4a804be11c36ee97a1dd75cbe4740fff3b534f8c` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / spanish](https://benchize.com/benchmarks/lmarena%3Atext%3Aspanish.md) | 1,523.42 | — | 1 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1523.4248373523887 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1503.7064897083796 rank 1 upper 1543.1431849963974 variance 101.21499460450697 votes 993 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10765/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate.md) | 87.7% | SE ±0.4 pp | 18 / 49 | $23.67 | 3,338.27source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.697 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.669537 \| USD \| per_test \| Lower is better \| \| `latency` \| 3338.269 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.4 pp sourcePopulation 81 stderr 0.395 tokenTotals {"cache_read_tokens":2623329853,"cache_write_tokens":134339560,"input_tokens":2757683085,"output_tokens":27559556,"reasoning_tokens":6673656} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [SRE Bench / overall](https://benchize.com/benchmarks/vals-ai%3Asrebench%3Aoverall.md) | 12.21% | SE ±2.03 pp | 9 / 24 | $23.63 | 4,858.21source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 12.214 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.633213 \| USD \| per_test \| Lower is better \| \| `latency` \| 4858.213 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.03 pp sourcePopulation 35 stderr 2.027 tokenTotals {"cache_read_tokens":6982893726,"cache_write_tokens":208290816,"input_tokens":7191259372,"output_tokens":50258561,"reasoning_tokens":34672581} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:15.724Z Evidence SHA-256 `a10b2c964f5878b0fc21c888703f311a5ef7d7a74fc3c529cab3ddd9fa474e2b` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,512.29 | — | 8 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1512.2895453084454 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1499.6233066234695 rank 9 upper 1524.9557839934212 variance 41.76371787731097 votes 2545 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6321/row` |  |  |  |  |  |  |  |
| [MMLU Pro / physics](https://benchize.com/benchmarks/vals-ai%3Ammlu_pro%3Aphysics.md) | 94.77% | SE ±0.62 pp | 3 / 40 | $0.03 | 9source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.765 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.025532 \| USD \| per_test \| Lower is better \| \| `latency` \| 8.999 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.62 pp sourcePopulation 138 stderr 0.618 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmlu_pro.y416pXDi.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:06.503Z Evidence SHA-256 `dab025507f46c716dd8d145e90ec0a737d9c56219853f41e4b1cc3c5e8c222b4` Source locator `/tasks/physics/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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 | $31.45 | 4,101.84source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 31.449565 \| USD \| per_test \| Lower is better \| \| `latency` \| 4101.838 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/data_privacy_cybersecurity/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,525.42 | — | 3 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1525.4151156615462 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1512.863258813179 rank 3 upper 1537.9669725099138 variance 41.01283332602265 votes 2545 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6315/row` |  |  |  |  |  |  |  |
| [Vals Science Index / overall](https://benchize.com/benchmarks/vals-ai%3Ascience_index%3Aoverall.md) | 47.93% | SE ±2.2 pp | 6 / 17 | $13.04 | 3,705.65source duration | [Vals AI · 1.0](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.93 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 13.038951 \| USD \| per_test \| Lower is better \| \| `latency` \| 3705.653 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.2 pp sourcePopulation 21 stderr 2.199 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1.0 Observed 2026-10-21T16:45:50.879Z Evidence SHA-256 `fa4313008e91752fcfc185aaa78665f780d0fc23d7c7198a9853dbf2eaaaef1f` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / industry writing and literature and language](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_writing_and_literature_and_language.md) | 1,480.68 | — | 14 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1480.6847375967081 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1473.452876913064 rank 14 upper 1487.9165982803524 variance 13.614569721767221 votes 9780 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7131/row` |  |  |  |  |  |  |  |
| [Arena webdev / overall](https://benchize.com/benchmarks/lmarena%3Awebdev%3Aoverall.md) | 1,688.86 | — | 7 / 90 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1688.8555367369042 \| rating \| reported \| Higher is better \| ### Configuration category overall config webdev Reasoning effort max thinkingEnabled true ### Reported statistics lower 1682.6517812580007 rank 8 upper 1695.0592922158078 variance 10.01874127471985 votes 18555 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/webdev/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `f82507859808110820a4ab6df15561b8864137ddefc1f6fb0c552af2e58ea738` Source locator `/rows/7/row` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / science](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Ascience.md) | 47.62% | SE ±2.38 pp | 7 / 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` \| 47.619 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.38 pp sourcePopulation 52 stderr 2.381 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/science/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [BioMysteryBench / human solvable](https://benchize.com/benchmarks/vals-ai%3Abiomysterybench%3Ahuman_solvable.md) | 87.67% | SE ±2.09 pp | 4 / 20 | $2.64 | 2,195.96source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.671 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.64447 \| USD \| per_test \| Lower is better \| \| `latency` \| 2195.964 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Terminus 2 maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.09 pp sourcePopulation 27 stderr 2.093 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:04.029Z Evidence SHA-256 `59bf79d38ec415a34ffb1c2901a16ecbbe101b7fa864e86104d2571fc7cfc23d` Source locator `/tasks/human_solvable/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Humanity's Sixth Sense](https://benchize.com/benchmarks/scale%3Ahss.md) | 30.5source score | CI 27–34 | 15 / 24 | — | — | [scale](https://labs.scale.com/leaderboard/hss) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 30.5 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort max Model configuration Claude-Opus-5 (max) sourceModelVersion Not reported thinkingEnabled true ### Reported statistics Uncertainty Confidence interval: 27–34 contaminationWarning Not reported rank 16 reportedConfidenceHalfWidth 3.5 ### Source [Original source](https://labs.scale.com/leaderboard/hss) Source scale Original model label Claude-Opus-5 (max) Observed 2026-10-21T16:35:26.953Z Evidence SHA-256 `0b7d37655e8886c99f8325ed125eafaa4f89312cd44e1a783ac0c49b45db842b` Source locator `/rows/15` |  |  |  |  |  |  |  |
| [CyberBench v1.1 / patch](https://benchize.com/benchmarks/benchmark%3Acyber%3Apatch.md) | 85.71% | SE ±4.68 pp | 11 / 28 | $2.01 | 1,230.08source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 85.714 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.005228 \| USD \| per_test \| Lower is better \| \| `latency` \| 1230.084 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.68 pp sourcePopulation 47 stderr 4.676 tokenTotals {"cache_read_tokens":187975583,"cache_write_tokens":5014307,"input_tokens":192995538,"output_tokens":1802901,"reasoning_tokens":1157748} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1.1 Observed 2026-10-21T16:46:15.238Z Evidence SHA-256 `943f593f929f8971a6b746269a2851b5febc2aad4e2f1066895f715033cdc7f1` Source locator `/tasks/patch/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / media](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Amedia.md) | 58.33% | SE ±8.33 pp | 6 / 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` \| 58.333 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.33 pp sourcePopulation 52 stderr 8.333 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/media/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate data privacy cybersecurity](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__data_privacy_cybersecurity.md) | 94.35% | SE ±1.38 pp | 12 / 49 | $31.45 | 4,101.84source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 94.346 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 31.449565 \| USD \| per_test \| Lower is better \| \| `latency` \| 4101.838 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.38 pp sourcePopulation 81 stderr 1.375 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__data_privacy_cybersecurity/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass current temporal analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__current_temporal_analysis.md) | 71.63% | SE ±8.05 pp | 12 / 46 | $6.01 | 1,207.61source 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` \| 71.631 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.007301 \| USD \| per_test \| Lower is better \| \| `latency` \| 1207.608 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.05 pp sourcePopulation 73 stderr 8.048 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__current_temporal_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text style control / creative writing](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acreative_writing.md) | 1,472.67 | — | 14 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1472.6651026949667 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text_style_control Reasoning effort max thinkingEnabled true ### Reported statistics lower 1464.5725742205314 rank 14 upper 1480.7576311694024 variance 17.04795500000045 votes 7695 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/1235/row` |  |  |  |  |  |  |  |
| [SRE Bench / partial](https://benchize.com/benchmarks/vals-ai%3Asrebench%3Apartial.md) | 31.11% | SE ±2.31 pp | 9 / 24 | $23.63 | 4,858.21source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 31.107 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.633213 \| USD \| per_test \| Lower is better \| \| `latency` \| 4858.213 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.31 pp sourcePopulation 35 stderr 2.314 tokenTotals {"cache_read_tokens":6982893726,"cache_write_tokens":208290816,"input_tokens":7191259372,"output_tokens":50258561,"reasoning_tokens":34672581} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:15.724Z Evidence SHA-256 `a10b2c964f5878b0fc21c888703f311a5ef7d7a74fc3c529cab3ddd9fa474e2b` Source locator `/tasks/partial/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Legal Research Bench / overall](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aoverall.md) | 55.29% | SE ±3.46 pp | 1 / 50 | $6.76 | 1,860.72source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 55.288 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.762498 \| USD \| per_test \| Lower is better \| \| `latency` \| 1860.719 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.46 pp sourcePopulation 82 stderr 3.456 tokenTotals {"cache_read_tokens":540621000,"cache_write_tokens":86242293,"input_tokens":626960894,"output_tokens":23871466,"reasoning_tokens":9873124} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [IOI / overall](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aoverall.md) | 84.33% | SE ±9.96 pp | 8 / 29 | $16.48 | 3,402.05source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.333 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 16.477241 \| USD \| per_test \| Lower is better \| \| `latency` \| 3402.045 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.96 pp sourcePopulation 44 stderr 9.964 tokenTotals {"cache_read_tokens":93265146,"cache_write_tokens":2811668,"input_tokens":96077494,"output_tokens":1386182,"reasoning_tokens":1125999} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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) | 93.91% | SE ±1.36 pp | 7 / 49 | $26 | 3,391.35source 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` \| 93.91 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 25.998286 \| USD \| per_test \| Lower is better \| \| `latency` \| 3391.348 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.36 pp sourcePopulation 81 stderr 1.356 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_ma/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Legal Research Bench / constitutional civil rights](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aconstitutional_civil_rights.md) | 44.74% | SE ±8.17 pp | 4 / 50 | $6.4 | 1,898.74source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 44.737 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.396226 \| USD \| per_test \| Lower is better \| \| `latency` \| 1898.737 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.17 pp sourcePopulation 82 stderr 8.174 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/constitutional_civil_rights/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [LegalBench / rule tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Arule_tasks.md) | 84.81% | SE ±1.48 pp | 13 / 44 | $0.02 | 11.2source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 84.805 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.020142 \| USD \| per_test \| Lower is better \| \| `latency` \| 11.197 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.48 pp sourcePopulation 151 stderr 1.477 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.390Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/rule_tasks/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [MMMU Pro / overall](https://benchize.com/benchmarks/vals-ai%3Ammmu%3Aoverall.md) | 89.88% | SE ±0.73 pp | 2 / 28 | $0.07 | 30.03source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mmmu.BBln8x5x.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.884 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.069014 \| USD \| per_test \| Lower is better \| \| `latency` \| 30.028 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.73 pp sourcePopulation 93 stderr 0.725 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mmmu.BBln8x5x.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:07.000Z Evidence SHA-256 `29395186eddd2c70601228c584a14b3c1d8f9dba4b68256561b46b0dd03c635a` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate trusts estates private client](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__trusts_estates_private_client.md) | 86.64% | SE ±2.11 pp | 24 / 49 | $24.98 | 3,106.68source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.641 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.97725 \| USD \| per_test \| Lower is better \| \| `latency` \| 3106.68 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.11 pp sourcePopulation 81 stderr 2.106 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__trusts_estates_private_client/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Code Migration / overall](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Aoverall.md) | 57.47% | SE ±4.37 pp | 7 / 47 | $60.51 | 10,268.73source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 57.473 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 60.509587 \| USD \| per_test \| Lower is better \| \| `latency` \| 10268.729 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.37 pp sourcePopulation 79 stderr 4.371 tokenTotals {"cache_read_tokens":13778689514,"cache_write_tokens":170020888,"input_tokens":13948783938,"output_tokens":48168536,"reasoning_tokens":null} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/overall/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / white collar defense investigations](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Awhite_collar_defense_investigations.md) | 0% | SE ±0 pp | 1 / 49 | $16.15 | 2,613.81source 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` \| 16.147201 \| USD \| per_test \| Lower is better \| \| `latency` \| 2613.808 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/white_collar_defense_investigations/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / engineering sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aengineering_sciences.md) | 11.11% | SE ±11.11 pp | 7 / 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` \| 11.111 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Mini-SWE-agent maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±11.11 pp sourcePopulation 41 stderr 11.111 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 anthropic/claude-opus-5 Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/engineering_sciences/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate intellectual property](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__intellectual_property.md) | 91.52% | SE ±1.66 pp | 17 / 49 | $23.47 | 3,194.55source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.519 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 23.467391 \| USD \| per_test \| Lower is better \| \| `latency` \| 3194.55 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.66 pp sourcePopulation 81 stderr 1.659 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__intellectual_property/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / industry legal and government](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_legal_and_government.md) | 1,529.67 | — | 2 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1529.6658035241644 \| rating \| reported \| Higher is better \| ### Configuration category industry_legal_and_government config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1517.9588229734086 rank 2 upper 1541.3727840749202 variance 35.677434019977056 votes 2901 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5109/row` |  |  |  |  |  |  |  |
| [Tax Agent Bench / current temporal analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Acurrent_temporal_analysis.md) | 49.88% | SE ±9.26 pp | 9 / 46 | $6.01 | 1,207.61source 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` \| 49.879 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.007301 \| USD \| per_test \| Lower is better \| \| `latency` \| 1207.608 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.26 pp sourcePopulation 73 stderr 9.263 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/current_temporal_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / market analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Amarket_analysis.md) | 74.16% | SE ±1.47 pp | 4 / 50 | $2.83 | 501.72source 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` \| 74.16 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.827226 \| USD \| per_test \| Lower is better \| \| `latency` \| 501.718 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.47 pp sourcePopulation 80 stderr 1.471 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/market_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / japanese](https://benchize.com/benchmarks/lmarena%3Atext%3Ajapanese.md) | 1,520.11 | — | 3 / 120 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1520.1053462024583 \| rating \| reported \| Higher is better \| ### Configuration category japanese config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1493.4554774806086 rank 3 upper 1546.7552149243083 variance 184.8817170877544 votes 594 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7953/row` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / life sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Alife_sciences.md) | 31.58% | SE ±10.96 pp | 7 / 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` \| 31.579 \| percent \| reported \| Higher is better \| ### Configuration computeEffort max harness Mini-SWE-agent maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±10.96 pp sourcePopulation 41 stderr 10.956 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 anthropic/claude-opus-5 Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/life_sciences/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass controversy precedence analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__controversy_precedence_analysis.md) | 70.02% | SE ±7.88 pp | 4 / 46 | $5.66 | 1,182.24source 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` \| 70.021 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.663294 \| USD \| per_test \| Lower is better \| \| `latency` \| 1182.24 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.88 pp sourcePopulation 73 stderr 7.882 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__controversy_precedence_analysis/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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) | 93% | SE ±1.38 pp | 10 / 49 | $28.96 | 3,945.16source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.003 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 28.957365 \| USD \| per_test \| Lower is better \| \| `latency` \| 3945.164 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.38 pp sourcePopulation 81 stderr 1.379 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__real_estate/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [IOI / ioi2025](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aioi2025.md) | 87.17% | SE ±11.72 pp | 7 / 29 | $16.52 | 3,406.43source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.167 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 16.518293 \| USD \| per_test \| Lower is better \| \| `latency` \| 3406.434 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±11.72 pp sourcePopulation 44 stderr 11.715 tokenTotals {"cache_read_tokens":94974943,"cache_write_tokens":2613707,"input_tokens":97589394,"output_tokens":1411316,"reasoning_tokens":1151620} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/ioi2025/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [SAGE / calculus](https://benchize.com/benchmarks/vals-ai%3Asage%3Acalculus.md) | 77.16% | SE ±5.31 pp | 11 / 36 | $0.52 | 58.45source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.16 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.524527 \| USD \| per_test \| Lower is better \| \| `latency` \| 58.451 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 30000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.31 pp sourcePopulation 91 stderr 5.312 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sage.RRwIIzO5.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:46:07.515Z Evidence SHA-256 `cc5fb3fecdb0382d492ff167a42ad84b9c5c3883373c78ed87baaa972b8b6f7c` Source locator `/tasks/calculus/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / creative writing](https://benchize.com/benchmarks/lmarena%3Atext%3Acreative_writing.md) | 1,493.45 | — | 6 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1493.4479039160842 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1485.3982980289024 rank 6 upper 1501.4975098032662 variance 16.867590663706327 votes 7695 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1227/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / corporate ma](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acorporate_ma.md) | 20% | SE ±20 pp | 1 / 49 | $26 | 3,391.35source 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` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 25.998286 \| USD \| per_test \| Lower is better \| \| `latency` \| 3391.348 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/corporate_ma/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Arena text / russian](https://benchize.com/benchmarks/lmarena%3Atext%3Arussian.md) | 1,507.13 | — | 8 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1507.1330768357548 \| rating \| reported \| Higher is better \| ### Configuration category russian config text Reasoning effort max thinkingEnabled true ### Reported statistics lower 1497.2374940872269 rank 8 upper 1517.0286595842824 variance 25.490982073124655 votes 4104 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Claude Opus 5 (Max) Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10391/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Abusiness_commercial.md) | 62.79% | SE ±7.46 pp | 1 / 50 | $7.36 | 2,163.79source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 62.791 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.357666 \| USD \| per_test \| Lower is better \| \| `latency` \| 2163.785 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.46 pp sourcePopulation 82 stderr 7.458 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/business_commercial/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Vals Multimodal Index / sage](https://benchize.com/benchmarks/vals-ai%3Avals_multimodal_index%3Asage.md) | 49.43% | SE ±3.29 pp | 10 / 20 | $0.19 | 56.17source duration | [Vals AI · 1.2](https://www.vals.ai/_astro/benchmark_view_vals_multimodal_index.virVpWMs.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 49.426 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.192726 \| USD \| per_test \| Lower is better \| \| `latency` \| 56.168 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.29 pp sourcePopulation 33 stderr 3.291 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_vals_multimodal_index.virVpWMs.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1.2 Observed 2026-10-21T16:45:51.881Z Evidence SHA-256 `077b22575f6e075b7b78157d000ec8f94d7ead409c7dcbcf04887630ed572a2d` Source locator `/tasks/sage/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [EMB / scratch](https://benchize.com/benchmarks/vals-ai%3Aemb%3Ascratch.md) | 72.26% | SE ±2.57 pp | 2 / 47 | $5.64 | 1,622.5source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.255 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.637053 \| USD \| per_test \| Lower is better \| \| `latency` \| 1622.496 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.57 pp sourcePopulation 76 stderr 2.573 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/scratch/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [CyberBench v1.1 / poc](https://benchize.com/benchmarks/benchmark%3Acyber%3Apoc.md) | 45% | SE ±6.42 pp | 18 / 28 | $3.73 | 1,100.72source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 45 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.731305 \| USD \| per_test \| Lower is better \| \| `latency` \| 1100.724 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.42 pp sourcePopulation 47 stderr 6.423 tokenTotals {"cache_read_tokens":308202284,"cache_write_tokens":4909598,"input_tokens":313117780,"output_tokens":2235503,"reasoning_tokens":1550456} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_cyberbench.DS6UOlu0.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1.1 Observed 2026-10-21T16:46:15.238Z Evidence SHA-256 `943f593f929f8971a6b746269a2851b5febc2aad4e2f1066895f715033cdc7f1` Source locator `/tasks/poc/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [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 | $24.31 | 3,388.15source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 24.310288 \| USD \| per_test \| Lower is better \| \| `latency` \| 3388.15 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±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 anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/banking_finance/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass family](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__family.md) | 86.65% | SE ±3.88 pp | 1 / 50 | $6.2 | 1,858.31source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.646 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 6.195885 \| USD \| per_test \| Lower is better \| \| `latency` \| 1858.312 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort max harness Not reported maxOutputTokens 128000 provider Anthropic reasoning Not reported Reasoning effort max Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.88 pp sourcePopulation 82 stderr 3.881 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label anthropic/claude-opus-5 Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__family/anthropic~1claude-opus-5` |  |  |  |  |  |  |  |
| [τ-bench banking knowledge](https://benchize.com/benchmarks/sierra%3Abanking_knowledge.md) | 48.71% | — | 1 / 19 | — | — | [sierra](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/claude-opus-5_sierra_2026-08-04/submission.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_1` \| 48.71134020618557 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_2` \| 39.175257731958766 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_3` \| 34.5360824742268 \| percent \| consistent_success_across_k_trials \| Higher is better \| \| `pass_4` \| 31.958762886597935 \| percent \| consistent_success_across_k_trials \| Higher is better \| ### Configuration benchmarkVersion 1.0.1 harness tau-bench methodology {"evaluation_date":"2026-08-03","notes":"Banking_knowledge evaluation with AllTools retrieval, GPT-5.2 low-reasoning user simulation, four trials, and seed 300.","tau2_bench_version":"1.0.1","user_simulator":"gpt-5.2","verification":{"modified_prompts":false,"omitted_questions":false}} modality text Reasoning effort max retrieval alltools sourceSubmission claude-opus-5_sierra_2026-08-04 ### Source [Original source](https://sierra-tau-bench-public.s3.us-west-2.amazonaws.com/submissions/claude-opus-5_sierra_2026-08-04/submission.json) Source sierra Original model label Claude Opus 5 Evaluated 2026-08-04 Observed 2026-10-21T16:54:53.721Z Evidence SHA-256 `4ab17f3f2e77cae8b2b4e4ba13413f7e2168104a772f25afe9d071950ed24ba2` Source locator `/results/banking_knowledge` |  |  |  |  |  |  |  |

[Next page](https://benchize.com/models/claude-opus-5-max.md?cursor=611a76df00bdd78923ff34ae52a235301b8cc2b53b6bf3b934811bb4389a498d)

---
Cite as: Claude Opus 5Max — Benchize, https://benchize.com/models/claude-opus-5-max (retrieved 2026-10-21)