Gemma 4 12B
Gemma 4 12B is a multimodal model developed by Google, capable of handling text, image, audio, and video inputs and generating text output, with a context window of up to 262,144 tokens. It is well-suited for tasks like text generation, coding, and reasoning, featuring both Dense and Mixture-of-Experts architectures.
Input
Output
Context
262K
Max Output
-
Parameters
12B
Technical Specifications
Model TypeChat
Context Window262,144 tokens
Max Output TokensNot available
Parameters12B
Release DateMay 23, 2026
Training CutoffJan 1, 2025
Licenseapache-2.0
Open SourceYes
Input Modalities
TextImageAudio
Output Modalities
TextImageAudio
Capabilities
Benchmarks
Artificial Analysis
14.8%
HLE
55.3%
AA-LCR
75.3%
GPQA
36.3%
Tau2-Bench
73.5%
IFBench
38.2%
SciCode
8.7%
Tau2 Banking
31.0
Coding
21.8
Intelligence
18.2%
Terminal-Bench Hard
27.3%
Terminal-Bench 2.1
126
Speed (tok/s)
17.3
TTFA (s)
1.4
TTFT (s)
Estimated GPU Requirements for Gemma 4 12B
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
More Models by Google
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Aug 8, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.