Human Frontier
80.6
Human-calibrated frontier signal, backed by Arena-style preference evidence and separate from raw AgMoBench benchmark composite scores.
Blended Price
$0.37/M
Input Price
$0.17/M
Output Price
$0.98/M
Speed
—
TTFT
—
Axes
Filters
Providers
vcontinuous-leaderboard · tasks-created:2025-08-01:2026-03-01 · GLM-4.5 Air · harness swe-rebench-react@tools · 5 attempts · pass@1 · swe-rebench:standardized:128k-or-model-limit · provider GLM-4.5 Air__tools · record GLM-4.5 Air__tools:1754006400000:1772323200000
{
"passAt5": 48.86363636363637,
"developer": "zhipu",
"instanceType": "model",
"contextPolicy": "128k unless the model supports less",
"attemptsPerTask": 5,
"totalTokenUsage": 1579348.8201062132,
"modelReleaseDate": "2025-07-28",
"scoreStandardError": 0.15542414370163868,
"cachedTokenPercentage": null,
"modelReleaseTimestamp": 1753660800000,
"standardContextTokens": 128000,
"averageInstanceCostUsd": 0.18917772252560847,
"potentialContamination": false,
"aggregateTaskRangeTimestamp": {
"to": 1772323200000,
"from": 1754006400000
},
"publishedTaskRangeTimestamp": {
"to": 1772323200000,
"from": 1754006400000
}
}Evidence observed 2026-03-01