工具条件未记录 / xhigh / 框架未记录 · FrontierCode v1.1 Main; mean of five runs per task; Cognition evaluation in Claude Code.
FrontierCode · v1.1 Main set · Overall score · 工具条件未记录 / xhigh / 框架未记录 · FrontierCode v1.1 Main; mean of five runs per task; Cognition evaluation in Claude Code.
工具条件未记录 / medium / Cognition Claude Code harness · FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.
FrontierCode · v1.1 Main set · Overall score · 工具条件未记录 / medium / Cognition Claude Code harness · FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.
Regular display scope. Rankings are grouped by metric and evaluation protocol. Recorded tool access, budgets, and frameworks are shown separately. Scores remain available for reference when conditions are unspecified.
Overall score %
Tool access not recorded / max / Framework not recorded · FrontierCode v1.1 Main。
FrontierCode · v1.1 Main set · Overall score · Tool access not recorded / max / Framework not recorded · FrontierCode v1.1 Main。
Tool access not recorded / xhigh / Framework not recorded · FrontierCode v1.1 Main; mean of five runs per task; Cognition evaluation in Claude Code.
FrontierCode · v1.1 Main set · Overall score · Tool access not recorded / xhigh / Framework not recorded · FrontierCode v1.1 Main; mean of five runs per task; Cognition evaluation in Claude Code.
Tool access not recorded / max / Framework not recorded · FrontierCode v1.1 Main, 150 software-engineering tasks.
FrontierCode · v1.1 Main set · Overall score · Tool access not recorded / max / Framework not recorded · FrontierCode v1.1 Main, 150 software-engineering tasks.
Tool access not recorded / xhigh / Framework not recorded · FrontierCode v1.1 Main, 150 software-engineering tasks.
FrontierCode · v1.1 Main set · Overall score · Tool access not recorded / xhigh / Framework not recorded · FrontierCode v1.1 Main, 150 software-engineering tasks.
Tool access not recorded / medium / Cognition Claude Code harness · FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.
FrontierCode · v1.1 Main set · Overall score · Tool access not recorded / medium / Cognition Claude Code harness · FrontierCode v1.1 Main; Cognition Claude Code harness; mean@5; medium effort.