Gemini 3 Flash

Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost. It combines Gemini 3's Pro-grade reasoning with Flash-level latency, efficiency and cost.

AIME 2025

99.7%

i
Global PIQA

92.8%

i
MMMLU

91.8%

i
GPQA

90.4%

i
t2-bench

90.2%

i
VideoMMMU

86.9%

i
MMMU-Pro

81.2%

i
CharXiv-R

80.3%

i
SWE-Bench Verified

78.0%

i
ScreenSpot Pro

69.1%

i
SimpleQA

68.7%

i
FACTS Grounding

61.9%

i
MCP Atlas

57.4%

i
Toolathlon

49.4%

i
Terminal-Bench 2.0

47.6%

i
Humanity's Last Exam

43.5%

i
ARC-AGI v2

33.6%

i
MRCR v2 (8-needle)

22.1%

i
OmniDocBench 1.5

12.1%

i
Vending-Bench 2

3,635

i
LiveCodeBench Pro

2,316

i

Pricing, uptime, and speed via OpenRouter — updated Jul 17, 2026, 04:19 AM.

Provider	Status	Input	Output	Limits	Uptime	Speed	Notes
Google	available	$0.90/Mtok cache $0.09/Mtok	$5.40/Mtok	1.0M tokens context 66K tokens max output	97% 5m 98%	2,478 ms p50 TTFT 70 tok/s p50	cache $0.000001/image · $0.014/web search
Google AI Studio	available	$0.90/Mtok cache $0.09/Mtok	$5.40/Mtok	1.0M tokens context 66K tokens max output	99.9% 5m 100.0%	—	cache $0.000001/image · $0.014/web search