gemma.cpp
The lightweight C++ inference engine for Google's Gemma model
CommonProductProgrammingMachine LearningC++
gemma.cpp is a lightweight and independent C++ inference engine developed by Google for the Gemma model. It focuses on providing a concise and direct implementation for the Gemma 2B and 7B models, aiming to support research and experimental cases, making it easy to integrate into other projects and offering high modularity. gemma.cpp utilizes the Google Highway library to optimize CPU inference performance.
gemma.cpp Visit Over Time
Monthly Visits
499904316
Bounce Rate
37.31%
Page per Visit
5.8
Visit Duration
00:06:52