Skip to main content
XOR

opus-4-6

run under the cursor harness

The harness is part of what is measured here: the same model under a different runner scores differently. Why that matters.

62.5%

repair rate

51–72

95% interval

2

rank of 15

$35.40

per repair · estimated

Rank is a position in a table whose intervals overlap almost everywhere. Read it as a band, not a place.

outcome mix

80 repaired · 24 not repaired · 24 did not build

80 repaired · 24 not repaired · 24 did not build

By weakness class

WeaknessRepair rateRateAttempts
CWE-125Out-of-bounds Read65.3% (95% interval 50.0–77.7)65.3%75
CWE-787Out-of-bounds Write76.5% (95% interval 58.8–93.8)76.5%17
CWE-416Use After Free46.2% (95% interval 20.0–87.5)46.2%13
CWE-908Use of Uninitialized Resource33.3% — no interval: too few codebases or attempts to generalise33.3%9
CWE-415Double Free66.7% — no interval: too few codebases or attempts to generalise66.7%3
CWE-1284Improper Validation of Specified Quantity in Input50.0% — no interval: too few codebases or attempts to generalise50.0%2
CWE-843Access of Resource Using Incompatible Type ('Type Confusion')50.0% — no interval: too few codebases or attempts to generalise50.0%2
CWE-476NULL Pointer Dereference50.0% — no interval: too few codebases or attempts to generalise50.0%2
CWE-475100.0% — no interval: too few codebases or attempts to generalise100.0%1
CWE-562100.0% — no interval: too few codebases or attempts to generalise100.0%1

Compare

vs codex/gpt-5-2vs claude/claude-opus-4-6vs gemini31/gemini-3-1-pro-previewvs opencode/gemini-3-1-pro-preview

Corpus cve-bench-136 · run February 2026 · How we test