New: GPT-6.2 benchmark report is here.Read the report ↗

Best LLMs on ExploitBench Internal Port

ExploitBench Internal Port

Develops exploits that achieve arbitrary code execution through recent V8 and Chrome vulnerabilities.

Updated: Data as of:
Limited reportingNot saturated
Official benchmark
ExploitBench Internal Port results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creatorVersionReported score
GPT-6.2 AstraOpenAI · ProprietaryJune–August 2026 / GPT-6.2 system card53.5%
GPT-6.2 SolOpenAI · ProprietaryJune–August 2026 / GPT-6.2 system card50.5%
View all 2 published reports
June–August 2026 / GPT-6.2 system card

aggregation: equal-weight mean · partial credit: false · targets: V8,Chrome

OpenAI ↗
June–August 2026 / GPT-6.2 system card

aggregation: equal-weight mean · partial credit: false · targets: V8,Chrome

OpenAI ↗