root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-25 02:44:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

scripts : fix typos, cleanup (#5303)

Browse Source
This commit is contained in:
Georgi Gerganov 2024-02-05 09:48:03 +02:00
parent 4be04c8965
commit 30679d438d
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 12 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
scripts/server-llm.sh
View File

@ -14,16 +14,17 @@
# - Might be unstable!
#
# Usage:
# ./server-llm.sh [--port] [--repo] [--wtype] [--backend] [--gpu-id] [--n-parallel] [--n-kv] [--verbose]
# ./server-llm.sh [--port] [--repo] [--wtype] [--backend] [--gpu-id] [--n-parallel] [--n-kv] [--verbose] [-non-interactive]
#
# --port: port number, default is 8888
# --repo: path to a repo containing GGUF model files
# --wtype: weights type (f16, q8_0, q4_0, q4_1), default is user-input
# --backend: cpu, cuda, metal, opencl, depends on the OS
# --gpu-id: gpu id, default is 0
# --n-parallel: number of parallel requests, default is 8
# --n-kv: KV cache size, default is 4096
# --verbose: verbose output
# --port: port number, default is 8888
# --repo: path to a repo containing GGUF model files
# --wtype: weights type (f16, q8_0, q4_0, q4_1), default is user-input
# --backend: cpu, cuda, metal, opencl, depends on the OS
# --gpu-id: gpu id, default is 0
# --n-parallel: number of parallel requests, default is 8
# --n-kv: KV cache size, default is 4096
# --verbose: verbose output
# --non-interactive: run without asking a permission to run
#
# Example:
#
@ -67,8 +68,7 @@ verbose=0
function print_usage {
printf "Usage:\n"
printf " ./server-llm.sh [-interactive] [--port] [--repo] [--wtype] [--backend] [--gpu-id] [--n-parallel] [--n-kv] [--verbose]\n\n"
printf " --non-interactive: run without asking a permision to run\n"
printf " ./server-llm.sh [--port] [--repo] [--wtype] [--backend] [--gpu-id] [--n-parallel] [--n-kv] [--verbose] [-non-interactive]\n\n"
printf " --port: port number, default is 8888\n"
printf " --repo: path to a repo containing GGUF model files\n"
printf " --wtype: weights type (f16, q8_0, q4_0, q4_1), default is user-input\n"
@ -77,6 +77,7 @@ function print_usage {
printf " --n-parallel: number of parallel requests, default is 8\n"
printf " --n-kv: KV cache size, default is 4096\n"
printf " --verbose: verbose output\n\n"
printf " --non-interactive: run without asking a permission to run\n"
printf "Example:\n\n"
printf ' bash -c "$(curl -s https://ggml.ai/server-llm.sh)"\n\n'
}