A low-latency, cost-effective natively multimodal Gemini model optimized for high-throughput agentic workflows and document parsing; it accepts text, image, video, audio, and PDF inputs and produces text.
Pricing
Model profile · Current snapshot
Input / 1M
$0.30
Output / 1M
$2.50
Blended / 1M
$0.85
Output TPS
403
TTFT (s)
4.77
Model Intelligence
Context window
1,048,576 tokens
Arena ranking
12
Benchmarkable
Yes
Model level
release
Intelligence Index
11.7
GPQA
0.84
HLE
0.19
SciCode
0.41
LCR
0.76
Recent stories
1 linked story