Overview
Designed for on-device applications.
Key Capabilities
●
On-device optimized
●
Native vision
●
Open weight
Benchmarks
MMLU Score
83%
HumanEval (Coding)
82%
GPQA Diamond
65%
MATH Benchmark
78%
Learn to Master This Model
Take our free structured Gemma course — from basics to advanced techniques.
Technical Specs
Context32,768 tokens
Params27B
LicenseGemma Terms
ArchTransformer
Developer
Google's AI research lab — creators of Gemini, the highest context-window multimodal LLM.
Prompt Library
Browse Coding Prompts →
Previous Version
Gemma 2 →