Human Frontier
92.2
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
vcontinuous-leaderboard · tasks-created:2025-10-01:2026-02-01 · Claude Opus 4.5 · harness swe-rebench-react@tools · 5 attempts · pass@1 · swe-rebench:standardized:128k-or-model-limit · provider Claude Opus 4.5__tools · record Claude Opus 4.5__tools:1759276800000:1769904000000
{
"passAt5": 65.4639175257732,
"developer": "anthropic",
"instanceType": "model",
"contextPolicy": "128k unless the model supports less",
"attemptsPerTask": 5,
"totalTokenUsage": 1358096.175257732,
"modelReleaseDate": "2025-11-24",
"scoreStandardError": 0.4312680549144704,
"cachedTokenPercentage": 95.173436766734,
"modelReleaseTimestamp": 1763942400000,
"standardContextTokens": 128000,
"averageInstanceCostUsd": 1.1468206239690724,
"potentialContamination": true,
"aggregateTaskRangeTimestamp": {
"to": 1769904000000,
"from": 1759276800000
},
"publishedTaskRangeTimestamp": {
"to": 1769904000000,
"from": 1759276800000
}
}Evidence observed 2026-02-01