jimmychen/NormalCommands
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

优化llama-cpp指令，添加host地址

Browse Source
This commit is contained in:
jimmychen
2026-03-29 00:12:55 -04:00
parent 900600e995
commit b8dbc4ac19
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llama-cpp.md
View File

@@ -1,4 +1,4 @@
./llama-server --model /home/jimmy/NVME/model/Qwen3.5-35B-A3B-UD-IQ4_XS.gguf --alias "Qwen3.5-35B-A3B" --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 --port 11434 --kv-unified --cache-type-k q8_0 --cache-type-v q8_0 --flash-attn on --fit on --ctx-size 262144 --jinja --no-mmap
./llama-server --model /home/jimmy/NVME/model/Qwen3.5-35B-A3B-UD-IQ4_XS.gguf --alias "Qwen3.5-35B-A3B" --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 --host 0.0.0.0 --port 11434 --kv-unified --cache-type-k q8_0 --cache-type-v q8_0 --flash-attn on --fit on --ctx-size 262144 --jinja --no-mmap -t 24 -np 4
context_size 128K = 131072