Frontier Log

OpenAIModel

GPT-5.4 mini

Made by
OpenAI
Released
Announced in
Introducing GPT-5.4 mini and nanoopenai.com
Event
GPT-5.4 mini and nano
API record
/api/v1/models/gpt-5-4-mini

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · xhigh effort28.2%
CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Humanity's Last ExamWith tools · xhigh effort; tools not named41.5%
CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
SWE-bench Proxhigh effort54.4%
CorroboratedThe same number is printed in 1 other lab document: 1
GPQA Diamondxhigh effort88.0%
CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Sources

Reported by OpenAI

Events

All models