Frontier Log

OpenAIModel

GPT-5.1

Made by
OpenAI
Released
Announced in
GPT-5.1: A smarter, more conversational ChatGPTopenai.com
Event
GPT-5.1
API record
/api/v1/models/gpt-5-1

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · reported later, in Introducing GPT-5.225.7%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
Humanity's Last ExamWith tools · search and Python · reported later, in Introducing GPT-5.242.7%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
SWE-bench Verifiedreported later, in Introducing GPT-5.276.3%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 6 other lab documents, the first 3 linked: 1 2 3
SWE-bench Proreported later, in Introducing GPT-5.250.8%
CorroboratedThe same number is printed in 1 other lab document: 1
GPQA Diamondreported later, in Introducing GPT-5.288.1%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 5 other lab documents, the first 3 linked: 1 2 3
BrowseCompreported later, in Introducing GPT-5.250.8%
CorroboratedThe same number is printed in 1 other lab document: 1
ARC-AGI-2ARC Prize Verified · reported later, in Introducing GPT-5.217.6%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 3 other lab documents: 1 2 3
Sources

Reported by OpenAI

Events

Stories

All models