Human Frontier
89.5
Human-calibrated frontier signal, backed by Arena-style preference evidence and separate from raw AgMoBench benchmark composite scores.
Blended Price
$4.00/M
Input Price
$2.00/M
Output Price
$10.00/M
Speed
66 tok/s
TTFT
6.73s
Axes
Filters
Providers
v2.1 · official-leaderboard · Sonnet 5 (high) · harness claude-code@2.1.205 · effort high · 5 attempts · harbor:terminal-bench/terminal-bench-2-1 · provider anthropic/claude-sonnet-5 · record leaderboard/submissions/2026-07-09-anthropic-claude-sonnet-5-high-claude-code.json · run 36288ba6-447b-5161-babf-cb46a228436c
{
"passAt2": 0.8652,
"passAt3": 0.8955,
"passAt4": 0.9101,
"passAt5": 0.9213,
"taskCount": 89,
"actualTokens": {
"output": 11216322,
"cachedInput": 527220403,
"uncachedInput": 20137729
},
"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/98",
"agentDisplayUrl": "https://claude.com/claude-code",
"modelDisplayUrl": "https://www.anthropic.com/news/claude-sonnet-5",
"actualTrialCount": 445,
"agentOrganization": "Anthropic",
"modelOrganization": "Anthropic",
"actualTotalCostUsd": 288.18,
"rewardHacksPercent": 0.67,
"accuracyStandardError": 1.64,
"submissionDownloadUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench-2-1/main/leaderboard/submissions/2026-07-09-anthropic-claude-sonnet-5-high-claude-code.json",
"disqualifiedTrialCount": 3,
"expectedAttemptsPerTask": 5,
"sourceFilterReasoningEffort": "high",
"actualAverageTrialDurationSeconds": 629.6
}Evidence observed 2026-07-09