Gemma 4 12B is Google's language model. A mid-size 12B-parameter open-weight language model from Google's Gemma 4 series, designed for efficient text generation and instruction following.
Specifications
Canonical IDgoogle-gemma-4-12b
TypeLanguage
StatusActive
CreatorGoogleGoogle
Input ModalitiesText
Output ModalitiesText
Parameters12B
Benchmarks
Coding Index
#105
GPQA
#219
HLE
#182
IFBench
#38
Time to First Token
#519
SciCode
#222
LCR
#196
TerminalBench Hard
#165
TAU2
#229
Output TPS
#120

Capabilities

Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Cheapest Instances to Run It

Cloud GPU instances that can host Gemma 4 12B, ranked by cheapest on-demand price. The model needs about 29 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.

All clouds
FP16 (full precision)
US Dollar ($)
Instance
Cloud
GPU
VRAM
Price
Cheapest region
Standard_NV16as_v4AzureAMD Radeon Instinct MI2532 GB$0.932/hr
Standard_NP10sAzureAMD Alveo U250 FPGA (64GB)64 GB$1.65/hr
g6e.xlargeAWSL40S45 GB$1.86/hr
7 more instances can run Gemma 4 12B
Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Gemma 4 12B————Current
Gemma 4 E4B————Available
Gemma 4 31B—256K$0.140$0.400Available
Gemma 4—262K$0.372$0.957Available
Gemma 4 12B Instruct—16K$0.300$2.00Available
Gemma 4 26B—131K$0.030$0.165Available
Gemma 4 26B A4B—262K$0.050$0.290Available
Gemma 4 26B A4B————Available
Gemma 4 31B————Available
Gemma 4 31B IT Thinking—262K$0.150$0.400Available
Gemma 4 31B IT Turbo—262K$0.090$0.340Available

Model IDs

gemma-4-12b
gemma-4-12b-non-reasoning
google-gemma-4-12b