Skip to main content
XOR

gpt-5-2

run under the opencode harness

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

51.6%

repair rate

40–63

95% interval

7

rank of 15

$5.96

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

63 repaired · 11 not repaired · 48 did not build

63 repaired · 11 not repaired · 48 did not build

By weakness class

WeaknessRepair rateRateAttempts
CWE-125Out-of-bounds Read62.5% (95% interval 45.5–75.3)62.5%72
CWE-787Out-of-bounds Write50.0% (95% interval 27.8–73.3)50.0%16
CWE-416Use After Free33.3% (95% interval 12.5–54.5)33.3%12
CWE-908Use of Uninitialized Resource22.2% — no interval: too few codebases or attempts to generalise22.2%9
CWE-415Double Free66.7% — no interval: too few codebases or attempts to generalise66.7%3
CWE-1284Improper Validation of Specified Quantity in Input0.0% — no interval: too few codebases or attempts to generalise0.0%2
CWE-476NULL Pointer Dereference0.0% — no interval: too few codebases or attempts to generalise0.0%2
CWE-843Access of Resource Using Incompatible Type ('Type Confusion')0.0% — no interval: too few codebases or attempts to generalise0.0%1
CWE-4750.0% — no interval: too few codebases or attempts to generalise0.0%1
CWE-5620.0% — no interval: too few codebases or attempts to generalise0.0%1

Compare

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

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