Human Frontier
90.3
Human-calibrated frontier signal, backed by Arena-style preference evidence and separate from raw AgMoBench benchmark composite scores.
Blended Price
$0.54/M
Input Price
$0.43/M
Output Price
$0.87/M
Speed
60 tok/s
TTFT
1.08s
Axes
Filters
Providers
vcontinuous-leaderboard · tasks-created:2026-03-01:2026-05-15 · DeepSeek-V4 Pro · harness swe-rebench-react@tools · 5 attempts · pass@1 · swe-rebench:standardized:128k-or-model-limit · provider DeepSeek-V4 Pro__tools · record DeepSeek-V4 Pro__tools:1772323200000:1778803200000
{
"passAt5": 61.81818181818181,
"developer": "deepseek",
"instanceType": "model",
"contextPolicy": "128k unless the model supports less",
"attemptsPerTask": 5,
"totalTokenUsage": 2247890.694713805,
"modelReleaseDate": "2026-04-24",
"scoreStandardError": 1.2913500456159093,
"cachedTokenPercentage": 76.93606904762608,
"modelReleaseTimestamp": 1776988800000,
"standardContextTokens": 128000,
"averageInstanceCostUsd": 0.2170780293797307,
"potentialContamination": true,
"aggregateTaskRangeTimestamp": {
"to": 1778803200000,
"from": 1772323200000
},
"publishedTaskRangeTimestamp": {
"to": 1778803200000,
"from": 1772323200000
}
}Evidence observed 2026-05-15