Difference between revisions of "LLaMA.cpp"
(Add stub for LLaMA.cpp) |
(Add programming languages (from llama-cpp f31b539 source code used in Guix)) |
||
| (2 intermediate revisions by 2 users not shown) | |||
| Line 4: | Line 4: | ||
|Full description=Port of Meta's LLaMA model in C/C++ to run on CPU instead of GPU. | |Full description=Port of Meta's LLaMA model in C/C++ to run on CPU instead of GPU. | ||
|Homepage URL=https://github.com/ggerganov/llama.cpp | |Homepage URL=https://github.com/ggerganov/llama.cpp | ||
| + | |Version download=https://github.com/ggerganov/llama.cpp/archive/refs/tags/b1266.tar.gz | ||
}} | }} | ||
{{Project license | {{Project license | ||
|License=Expat | |License=Expat | ||
| − | |License copyright=. | + | |License copyright=Copyright (c) 2023 Georgi Gerganov |
| − | |License verified by= | + | |
| − | |License verified date=2023- | + | The README.md inside the source code has: |
| + | [](https://opensource.org/licenses/MIT) | ||
| + | |||
| + | That is probably what makes "MIT license" to be displayed on the right of the page. | ||
| + | |||
| + | The files don't seem to have licensing headers but this license declaration is probably enough for the license to apply to all the files inside the source code. | ||
| + | |License verified by=GNUtoo, craigt | ||
| + | |License verified date=2023-09-22 | ||
| + | |License note=https://raw.githubusercontent.com/ggerganov/llama.cpp/master/LICENSE | ||
}} | }} | ||
{{Software category | {{Software category | ||
| + | |Programming-language=C, C++, python, javascript, shell | ||
| + | |Runs-on=Windows, BSD, OS X, GNU/Linux | ||
|Science=artificial-intelligence | |Science=artificial-intelligence | ||
}} | }} | ||
{{EntryGNU | {{EntryGNU | ||
|Is GNU=No | |Is GNU=No | ||
| + | }} | ||
| + | {{Software prerequisite | ||
| + | |Prerequisite kind=Required to use | ||
| + | |Prerequisite description=[[python]] | ||
| + | }} | ||
| + | {{Software prerequisite | ||
| + | |Prerequisite kind=Required to use | ||
| + | |Prerequisite description=[[numpy]] | ||
| + | }} | ||
| + | {{Software prerequisite | ||
| + | |Prerequisite kind=Required to use | ||
| + | |Prerequisite description=[[sentencepiece]] | ||
| + | }} | ||
| + | {{Software prerequisite | ||
| + | |Prerequisite kind=Required to use | ||
| + | |Prerequisite description=[[PyTorch]] | ||
}} | }} | ||
{{EntryDetails | {{EntryDetails | ||
|Is High Priority Project=No | |Is High Priority Project=No | ||
|VCS checkout command=git clone https://github.com/ggerganov/llama.cpp | |VCS checkout command=git clone https://github.com/ggerganov/llama.cpp | ||
| + | |Computer languages=C++,C,python,bash,sh,javascript | ||
|Decommissioned or Obsolete=No | |Decommissioned or Obsolete=No | ||
|Accepts cryptocurrency donations=No | |Accepts cryptocurrency donations=No | ||
Latest revision as of 06:31, 24 September 2023
LLaMA.cpp
https://github.com/ggerganov/llama.cpp
Inference of LLaMA model in pure C/C++
Port of Meta's LLaMA model in C/C++ to run on CPU instead of GPU.
Download
https://github.com/ggerganov/llama.cpp/archive/refs/tags/b1266.tar.gz
VCS Checkout
git clone https://github.com/ggerganov/llama.cppCategories
Licensing
License
Verified by
Verified on
Notes
License
Verified by
GNUtoo, craigt
Verified on
22 September 2023
Leaders and contributors
Resources and communication
Software prerequisites
| Kind | Description |
|---|---|
| Required to use | sentencepiece |
| Required to use | python |
| Required to use | PyTorch |
| Required to use | numpy |
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the page “GNU Free Documentation License”.
The copyright and license notices on this page only apply to the text on this page. Any software or copyright-licenses or other similar notices described in this text has its own copyright notice and license, which can usually be found in the distribution or license text itself.