google
Gemini 3.7 Flash
Gemini 3.7 Flash is the high-efficiency, cost-effective powerhouse of the Gemini 3 family. It delivers Pro-level agentic capabilities, major leaps in code generation and terminal execution. 3.7 Flash serves as the primary agentic workhorse in the Gemini 3 family, bridging the gap between deep-reasoning Pro models and high-throughput Flash-Lite models while delivering high token efficiency and multi-step multimodal processing.
reasoningfile inputvisiontool useweb searchimplicit cachingvideo inputstructured outputtextimagepdfvideo
Overview
- Type
- language
- Context
- 1M
- Max output
- 66K
- Input / output
- text, image, pdf, video / text
API
Model identifier
google/gemini-3.7-flashProviders and live performance
Rolling Gateway metrics. Latency is time to first token; throughput is output tokens per second.
| Provider | Status | Latency | Throughput | Uptime 15m / 1h / 1d | Input | Output | Cache read / write | Discount | Regions | ZDR | No training |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Operational | 1.48s / 7.91s | 139.0 / 186.0 tok/s | 100.00% / 99.65% / 99.84% | $0.75/M | $3.75/M | $0.07/M / — | — | — | — | Yes | |
| vertex | Operational | 2.43s / 13.37s | 151.0 / 229.4 tok/s | 100.00% / 99.99% / 99.96% | $0.75/M | $3.75/M | $0.07/M / — | — | eu, us | Yes | Yes |
Pricing
input
0.00000075
output
0.00000375
input cache read
0.000000075
web search
14
maps search
14
service tiers
{
"flex": {
"input": "0.000000375",
"output": "0.000001875",
"input_cache_read": "0.0000000375"
},
"priority": {
"input": "0.00000135",
"output": "0.00000675",
"input_cache_read": "0.000000135"
}
}
regional
{
"eu": {
"input": "0.000000825",
"output": "0.000004125",
"input_cache_read": "0.0000000825"
},
"us": {
"input": "0.000000825",
"output": "0.000004125",
"input_cache_read": "0.0000000825"
}
}
Capabilities
reasoningfile inputvisiontool useweb searchimplicit cachingvideo inputstructured outputtextimagepdfvideo
Supported parameters
max tokens
Yes
temperature
Yes
stop
Yes
tools
Yes
tool choice
Yes
reasoning
Yes
include reasoning
Yes
response format
Yes
structured outputs
Yes
Reasoning options
options
{
"type": "effort",
"values": [
"low",
"high"
]
}
Metadata
type
language
context window
1M
max output
66K
knowledge cutoff
2026-03
released
Aug 13, 2026
temperature control
Yes
zero data retention
some
no training
all
regions
eu, us
api specifications
v2, v3, v4