Frontier Log

GoogleModel

Gemini 3.5 Flash

Made by
Google
Released
Announced in
Gemini 3.5: frontier intelligence with actionblog.google
Event
Gemini 3.5 Flash
API record
/api/v1/models/gemini-3-5-flash

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · full set, text and multimodal40.2%
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 Prosingle attempt55.1%
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 2 other lab documents: 1 2
ARC-AGI-272.1%
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.
DeepSWE v1.1reported later, in Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber37%
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
Sources

Reported by Google

Events

Stories

All models