lightweight, standalone C++ inference engine for Google's Gemma models.
Go to file
prajwalc22 01caf379ba Update .gitignore to exclude build directory and model files 2025-04-16 09:45:14 +05:30
.github/workflows Update github actions/cache version 2025-03-12 07:12:55 -07:00
.vscode Add comments regarding layers output usage. 2024-04-12 15:39:16 +00:00
backprop Major refactor of allocator/args: 2025-04-10 01:29:54 -07:00
bazel Internal changes 2024-11-13 09:01:38 -08:00
compression Add mmap support (not yet used) 2025-04-10 10:03:40 -07:00
docs initial commit 2024-02-21 03:31:22 +00:00
evals Major refactor of allocator/args: 2025-04-10 01:29:54 -07:00
examples Major refactor of allocator/args: 2025-04-10 01:29:54 -07:00
experimental Update README with more details on contributing code, add experimental/ directory, add READMEs for subdirectories, clean up DEVELOPER notes 2024-03-15 14:10:24 -04:00
gemma Add --prompt flag for non-interactive mode 2025-04-16 09:45:02 +05:30
goldens Add a benchmark and additional tests. 2024-04-06 12:54:52 +02:00
ops Major refactor of allocator/args: 2025-04-10 01:29:54 -07:00
paligemma Add mmap support (not yet used) 2025-04-10 10:03:40 -07:00
python Add mmap support (not yet used) 2025-04-10 10:03:40 -07:00
util Add mmap support (not yet used) 2025-04-10 10:03:40 -07:00
.bazelrc Support Bazel builds. Fixes #16 2024-03-04 22:07:25 -08:00
.bazelversion .bazelversion: Bazel 7.1.1 2024-03-31 11:39:21 -07:00
.clang-format Add clang-tidy, fix narrowing issues, fix constness 2024-02-28 20:04:09 +08:00
.clang-tidy Add a benchmark and additional tests. 2024-04-06 12:54:52 +02:00
.gitignore Update .gitignore to exclude build directory and model files 2025-04-16 09:45:14 +05:30
BUILD.bazel Add mmap support (not yet used) 2025-04-10 10:03:40 -07:00
CMakeLists.txt Update CMake configuration and documentation for --prompt flag 2025-04-16 09:45:14 +05:30
CMakePresets.json Restore reverted changes. 2024-02-25 19:32:07 -08:00
DEVELOPERS.md Cleanup: move util/compress and convert_weights to compression/ 2024-07-05 04:16:52 -07:00
LICENSE initial commit 2024-02-21 03:31:22 +00:00
LICENSE-BSD3 initial commit 2024-02-21 03:31:22 +00:00
MODULE.bazel Refactor Gemma ctor and improve pool NUMA support 2025-03-14 10:19:00 -07:00
README.md Update CMake configuration and documentation for --prompt flag 2025-04-16 09:45:14 +05:30
WORKSPACE Support Bazel builds. Fixes #16 2024-03-04 22:07:25 -08:00
cmake.sh Add a benchmark and additional tests. 2024-04-06 12:54:52 +02:00

README.md

library_name license pipeline_tag tags extra_gated_heading extra_gated_prompt extra_gated_button_content
gemma.cpp gemma text-generation
Access Gemma on Hugging Face To access Gemma on Hugging Face, youre required to review and agree to Googles usage license. To do this, please ensure youre logged-in to Hugging Face and click below. Requests are processed immediately. Acknowledge license

Gemma Model Card

Model Page: Gemma

This model card corresponds to the 2B base version of the Gemma model for usage with C++ (https://github.com/google/gemma.cpp). This is a compressed version of the weights, which will load, run, and download more quickly. For more information about the model, visit https://huggingface.co/google/gemma-2b.

Resources and Technical Documentation:

Terms of Use: Terms

Authors: Google