Anthropic's most capable generally available model. Step-change in agentic coding, stronger vision (3.75MP), long-context reasoning, and knowledge work.
Human Frontier
95.5
Human-calibrated frontier signal, backed by Arena-style preference evidence and separate from raw AgMoBench benchmark composite scores.
Blended Price
$10.00/M
Input Price
$5.00/M
Output Price
$25.00/M
Speed
—
TTFT
—
Axes
Filters
Providers
v3 · public-demo · Opus 4.7 (High) · effort high · 1 attempt · 5x-human-baseline-median-per-level · arc-agi-3:public-demo · record v3_Semi_Private:anthropic-opus-4-7-high
{
"provider": "Anthropic",
"modelType": "CoT",
"modelsUrl": "https://arcprize.org/media/data/models.json",
"modelGroup": "anthropic-opus-4-7",
"modelNotes": null,
"modelsEtag": "W/\"2fc19b6caba8b52765ca17eab04fc437\"",
"scoreMetric": "relative-human-action-efficiency",
"evaluationsUrl": "https://arcprize.org/media/data/evaluations.json",
"evaluationsEtag": "W/\"9485c7ef206d9b12cdfdec6a631df56b\"",
"upstreamModelId": "anthropic-opus-4-7-high",
"environmentCount": 25,
"modelReleaseDate": "2026-04-16T00:00:00.000Z",
"upstreamDatasetId": "v3_Semi_Private",
"actionBudgetPolicy": "5x-human-baseline-median-per-level",
"technicalReportUrl": "https://arcprize.org/media/ARC_AGI_3_Technical_Report.pdf",
"catalogMatchAliases": [
"claude-opus-4-7"
],
"agentHarnessPublished": false,
"verificationAuthority": "ARC Prize Foundation",
"upstreamRunIdPublished": false,
"actualEvaluationCostUsd": 10000,
"providerModelIdPublished": false
}Evidence observed 2026-07-31
v3 · public-demo · Opus 4.7 (High) · effort high · 1 attempt · 5x-human-baseline-median-per-level · arc-agi-3:public-demo · record v3_Semi_Private:anthropic-opus-4-7-high
{
"provider": "Anthropic",
"modelType": "CoT",
"modelsUrl": "https://arcprize.org/media/data/models.json",
"modelGroup": "anthropic-opus-4-7",
"modelNotes": null,
"modelsEtag": "W/\"2fc19b6caba8b52765ca17eab04fc437\"",
"scoreMetric": "relative-human-action-efficiency",
"evaluationsUrl": "https://arcprize.org/media/data/evaluations.json",
"evaluationsEtag": "W/\"9485c7ef206d9b12cdfdec6a631df56b\"",
"upstreamModelId": "anthropic-opus-4-7-high",
"environmentCount": 25,
"modelReleaseDate": "2026-04-16T00:00:00.000Z",
"upstreamDatasetId": "v3_Semi_Private",
"actionBudgetPolicy": "5x-human-baseline-median-per-level",
"technicalReportUrl": "https://arcprize.org/media/ARC_AGI_3_Technical_Report.pdf",
"catalogMatchAliases": [
"claude-opus-4-7"
],
"agentHarnessPublished": false,
"verificationAuthority": "ARC Prize Foundation",
"upstreamRunIdPublished": false,
"actualEvaluationCostUsd": 10000,
"providerModelIdPublished": false
}Evidence observed 2026-07-31
v3 · public-demo · Opus 4.7 (High) · effort high · 1 attempt · 5x-human-baseline-median-per-level · arc-agi-3:public-demo · record v3_Semi_Private:anthropic-opus-4-7-high
{
"provider": "Anthropic",
"modelType": "CoT",
"modelsUrl": "https://arcprize.org/media/data/models.json",
"modelGroup": "anthropic-opus-4-7",
"modelNotes": null,
"modelsEtag": "W/\"cfc9b87c645d3ef2103e359fd0f9f521\"",
"scoreMetric": "relative-human-action-efficiency",
"evaluationsUrl": "https://arcprize.org/media/data/evaluations.json",
"evaluationsEtag": "W/\"0d65f6e0efe0d5cae15e8e2fe5a0994b\"",
"upstreamModelId": "anthropic-opus-4-7-high",
"environmentCount": 25,
"modelReleaseDate": "2026-04-16T00:00:00.000Z",
"upstreamDatasetId": "v3_Semi_Private",
"actionBudgetPolicy": "5x-human-baseline-median-per-level",
"technicalReportUrl": "https://arcprize.org/media/ARC_AGI_3_Technical_Report.pdf",
"catalogMatchAliases": [
"claude-opus-4-7"
],
"agentHarnessPublished": false,
"verificationAuthority": "ARC Prize Foundation",
"upstreamRunIdPublished": false,
"actualEvaluationCostUsd": 10000,
"providerModelIdPublished": false
}Evidence observed 2026-07-30
v3 · public-demo · Opus 4.7 (High) · effort high · 1 attempt · 5x-human-baseline-median-per-level · arc-agi-3:public-demo · record v3_Semi_Private:anthropic-opus-4-7-high
{
"provider": "Anthropic",
"modelType": "CoT",
"modelsUrl": "https://arcprize.org/media/data/models.json",
"modelGroup": "anthropic-opus-4-7",
"modelNotes": null,
"modelsEtag": "W/\"bbdf5725ef7a703a7f0970e913262b6f\"",
"scoreMetric": "relative-human-action-efficiency",
"evaluationsUrl": "https://arcprize.org/media/data/evaluations.json",
"evaluationsEtag": "W/\"6529094085f9ca7ce5f3c2e1957d3b9c\"",
"upstreamModelId": "anthropic-opus-4-7-high",
"environmentCount": 25,
"modelReleaseDate": "2026-04-16T00:00:00.000Z",
"upstreamDatasetId": "v3_Semi_Private",
"actionBudgetPolicy": "5x-human-baseline-median-per-level",
"technicalReportUrl": "https://arcprize.org/media/ARC_AGI_3_Technical_Report.pdf",
"catalogMatchAliases": [
"claude-opus-4-7"
],
"agentHarnessPublished": false,
"verificationAuthority": "ARC Prize Foundation",
"upstreamRunIdPublished": false,
"actualEvaluationCostUsd": 10000,
"providerModelIdPublished": false
}Evidence observed 2026-07-24
v2026.06 · official-100-task · Claude Opus 4.7 (max) · harness mini-swe-agent · effort max · 1 attempt · pass@1 · harbor:snorkel-ai/senior-swe-bench-v2026.06 · provider anthropic/claude-opus-4-7 · record opus-mini-swe · run 1d69314975a9519f64c507748d04762ded7853ebf34982811a94ac3a27ba2f9e
{
"taskCount": 100,
"agentGroup": "mini-swe-agent:claude-opus-4-7",
"basicPassAt1": 39.08045977011494,
"datasetPackage": "snorkel-ai/senior-swe-bench-v2026.06",
"publicTaskCount": 50,
"tastefulPassAt1": 13.793103448275861,
"agentDescription": "Claude Opus 4.7 via mini-swe-agent harness (Portkey→Bedrock)",
"privateTaskCount": 50,
"upstreamProvider": "anthropic",
"upstreamRunCount": 50,
"averageAgentSteps": 157.95402298850576,
"averageOutputTokens": 97643.86206896552,
"averageTotalCostUsd": 8.871201140229886,
"averageOutputCostUsd": 2.4410965517241383
}Evidence observed 2026-06-30
vcontinuous-leaderboard · tasks-created:2026-03-01:2026-05-15 · Claude Opus 4.7-high · harness swe-rebench-react@tools · effort high · 5 attempts · pass@1 · swe-rebench:standardized:128k-or-model-limit · provider Claude Opus 4.7-high__tools · record Claude Opus 4.7-high__tools:1772323200000:1778803200000
{
"passAt5": 66.36363636363637,
"developer": "anthropic",
"instanceType": "model",
"contextPolicy": "128k unless the model supports less",
"attemptsPerTask": 5,
"totalTokenUsage": 1526135.3290909089,
"modelReleaseDate": "2026-04-16",
"scoreStandardError": 1.4545454545454541,
"cachedTokenPercentage": 94.18072169671689,
"modelReleaseTimestamp": 1776297600000,
"standardContextTokens": 128000,
"averageInstanceCostUsd": 1.3181382954545455,
"potentialContamination": true,
"aggregateTaskRangeTimestamp": {
"to": 1778803200000,
"from": 1772323200000
},
"publishedTaskRangeTimestamp": {
"to": 1778803200000,
"from": 1772323200000
}
}Evidence observed 2026-05-15
v2.1 · official-leaderboard · Opus 4.7 (max) · harness claude-code@2.1.123 · effort max · 5 attempts · harbor:terminal-bench/terminal-bench-2-1 · provider anthropic/claude-opus-4-7 · record leaderboard/submissions/2026-05-01-anthropic-claude-opus-4-7-max-claude-code.json · run 10e2e56b-ed31-5f65-a489-69f78b902adf
{
"passAt2": 0.7801,
"passAt3": 0.8124,
"passAt4": 0.83,
"passAt5": 0.8408,
"taskCount": 89,
"actualTokens": {
"output": 14896548,
"cachedInput": 806584422,
"uncachedInput": 31873388
},
"datasetPackage": "terminal-bench/terminal-bench-2-1",
"leaderboardUrl": "https://www.tbench.ai/leaderboard/terminal-bench/2.1",
"pullRequestUrl": "https://github.com/harbor-framework/terminal-bench-2-1/pull/44",
"agentDisplayUrl": "https://claude.com/claude-code",
"modelDisplayUrl": "https://www.anthropic.com/news/claude-opus-4-7",
"actualTrialCount": 447,
"agentOrganization": "Anthropic",
"modelOrganization": "Anthropic",
"actualTotalCostUsd": 599.52,
"rewardHacksPercent": 0.45,
"accuracyStandardError": 1.41,
"submissionDownloadUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench-2-1/main/leaderboard/submissions/2026-05-01-anthropic-claude-opus-4-7-max-claude-code.json",
"disqualifiedTrialCount": 2,
"expectedAttemptsPerTask": 5,
"sourceFilterReasoningEffort": "max",
"actualAverageTrialDurationSeconds": 799.6
}Evidence observed 2026-05-01
v2.1 · official-leaderboard · Opus 4.7 (max) · harness terminus-2@2.0.0 · effort max · 5 attempts · harbor:terminal-bench/terminal-bench-2-1 · provider anthropic/claude-opus-4-7 · record leaderboard/submissions/2026-05-01-anthropic-claude-opus-4-7-max-terminus-2.json · run 10e2e56b-ed31-5f65-a489-69f78b902adf
{
"passAt2": 0.7438,
"passAt3": 0.782,
"passAt4": 0.809,
"passAt5": 0.8315,
"taskCount": 89,
"actualTokens": {
"output": 12451178,
"cachedInput": 336161721,
"uncachedInput": 16646631
},
"datasetPackage": "terminal-bench/terminal-bench-2-1",
"leaderboardUrl": "https://www.tbench.ai/leaderboard/terminal-bench/2.1",
"pullRequestUrl": "https://github.com/harbor-framework/terminal-bench-2-1/pull/46",
"agentDisplayUrl": "https://www.tbench.ai/news/terminus",
"modelDisplayUrl": "https://www.anthropic.com/news/claude-opus-4-7",
"actualTrialCount": 445,
"agentOrganization": "Terminal-Bench",
"modelOrganization": "Anthropic",
"actualTotalCostUsd": 582.26,
"rewardHacksPercent": 0,
"accuracyStandardError": 1.37,
"submissionDownloadUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench-2-1/main/leaderboard/submissions/2026-05-01-anthropic-claude-opus-4-7-max-terminus-2.json",
"disqualifiedTrialCount": 0,
"expectedAttemptsPerTask": 5,
"sourceFilterReasoningEffort": "max",
"actualAverageTrialDurationSeconds": 1042.7
}Evidence observed 2026-05-01