commit 1ea57a11f46b6fb2b3f450206f6332d331cfead4 parent 5e118710cbcb512c83aed6d323d00169344d4806 Author: Oliver Lowe <oliver.lowe@audinate.com> Date: Sun, 10 Nov 2024 18:15:42 +1100 update README Diffstat:
| M | README | | | 2 | ++ |
| A | man/precis.1 | | | 21 | +++++++++++++++++++++ |
2 files changed, 23 insertions(+), 0 deletions(-)
diff --git a/README b/README @@ -4,11 +4,13 @@ To install, run install.sh. The following commands are provided: +- bbsnip - create a snippet on BitBucket - hits - count web traffic - hlsget - download the contents of a HLS playlist - jsfmt - format javascript source code - lemmyverse - find lemmy communities - llama - prompt a large language model +- precis - summarise text - webpaste - create a web paste on webpaste.olowe.co - ws - web search - xstream - stream X display over the network diff --git a/man/precis.1 b/man/precis.1 @@ -0,0 +1,21 @@ +.Dd +.Dt PRECIS 1 +.Sh NAME +.Nm precis +.Nd summarise text +.Sh SYNOPSIS +.Nm +.Op Ar model +.Sh DESCRIPTION +.Nm +reads text from the standard input +and prints a short summary using a large language model. +.Ar model +is a path to a gguf model file. +The default is +.Pa $HOME/llama-3.2-3b-instruct-q4_k_m.gguf . +.Sh EXIT STATUS +.Ex +.Sh SEE ALSO +.Xr llama-cli 1 , +.Lk https://github.com/ggerganov/llama.cpp llama.cpp