BenchLeader

Agnes 3.0 Flash vs Gemini 3.8 Flash

On the 13 benchmarks both have, Gemini 3.8 Flash scores higher on 10 and Agnes 3.0 Flash on 3.

No benchmark publisher we read lists Agnes 3.0 Flash any more, so its figures stopped updating on 22 Sep 2026. Gemini 3.8 Flash's figures are current as of 13 Nov 2026. Each score shows the date it was published.

BenchmarkAgnes 3.0 FlashGemini 3.8 Flash (high)
AA Intelligence Index v4.3.235.5
11 Sep 2026
40.9
2 Sep 2026
AA-LCR v1.181.0%
11 Sep 2026
81.3%
2 Sep 2026
AA-Omniscience v1.1-10.6
11 Sep 2026
23.3
2 Sep 2026
GPQA Diamond (AA)92.4%
11 Sep 2026
95.3%
2 Sep 2026
Humanity's Last Exam (AA)38.5%
11 Sep 2026
47.8%
2 Sep 2026
SciCode (AA)51.6%
11 Sep 2026
56.6%
2 Sep 2026
CritPt15.1%
19 Sep 2026
18.3%
2 Sep 2026
GDPval-AA v2.1, normalised score52.5%
20 Sep 2026
46.8%
2 Sep 2026
τ³-Banking (AA)47.6%
19 Sep 2026
45.0%
2 Sep 2026
Terminal-Bench 4.0 (AA)7.1%
22 Sep 2026
19.7%
2 Sep 2026
Terminal-Bench 2.1 (AA)82.0%
22 Sep 2026
87.6%
2 Sep 2026
AA-Omniscience v1.1: accuracy25.5%
22 Sep 2026
50.5%
2 Sep 2026
AA-Omniscience v1.1: non-hallucination51.6%
22 Sep 2026
44.9%
2 Sep 2026
Blended price (per 1M tokens, 3:1)$0.075$1.50
Output speed317 tok/s111 tok/s

See everything we last measured for Agnes 3.0 Flash, or Gemini 3.8 Flash's page for how it compares with models that are still measured. The full list is on the leaderboards.