localLLM: Running Local LLMs with 'llama.cpp' Backend
The 'localLLM' package provides R bindings to the 'llama.cpp' library for running large language models. The package uses a lightweight architecture where the C++ backend library is downloaded at runtime rather than bundled with the package. Package features include text generation, reproducible generation, and parallel inference.
| Version: | 1.0.1 |
| Depends: | R (≥ 3.6.0) |
| Imports: | Rcpp (≥ 1.0.14), tools, utils |
| Suggests: | testthat (≥ 3.0.0),covr |
| Published: | 2025-10-15 |
| DOI: | 10.32614/CRAN.package.localLLM |
| Author: | Eddie Yang [aut], Yaosheng Xu [aut, cre] |
| Maintainer: | Yaosheng Xu <xu2009 at purdue.edu> |
| BugReports: | https://github.com/EddieYang211/localLLM/issues |
| License: | MIT + fileLICENSE |
| URL: | https://github.com/EddieYang211/localLLM |
| NeedsCompilation: | yes |
| SystemRequirements: | C++17, libcurl (optional, for model downloading) |
| CRAN checks: | localLLM results |
Documentation:
| Reference manual: | localLLM.html ,localLLM.pdf |
Downloads:
| Package source: | localLLM_1.0.1.tar.gz |
| Windows binaries: | r-devel:localLLM_1.0.1.zip, r-release:localLLM_1.0.1.zip, r-oldrel:localLLM_1.0.1.zip |
| macOS binaries: | r-release (arm64):localLLM_1.0.1.tgz, r-oldrel (arm64):localLLM_1.0.1.tgz, r-release (x86_64):localLLM_1.0.1.tgz, r-oldrel (x86_64):localLLM_1.0.1.tgz |
Linking:
Please use the canonical formhttps://CRAN.R-project.org/package=localLLMto link to this page.