oobabooga/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2026-01-05 08:10:07 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

llama.cpp: don't include --rope-freq-base in the launch command if null

Browse source
This commit is contained in:
oobabooga 2025-04-18 20:51:18 -07:00
parent 9f07a1f5d7
commit e2e90af6cd
1 changed files with 2 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
modules/llama_cpp_server.py
View file

@ -250,7 +250,6 @@ class LlamaServer:
"--ctx-size", str(shared.args.n_ctx),
"--n-gpu-layers", str(shared.args.n_gpu_layers),
"--batch-size", str(shared.args.batch_size),
"--rope-freq-base", str(shared.args.rope_freq_base),
"--port", str(self.port),
]
@ -276,6 +275,8 @@ class LlamaServer:
cmd += ["--cache-type-k", shared.args.cache_type, "--cache-type-v", shared.args.cache_type]
if shared.args.compress_pos_emb != 1:
cmd += ["--rope-freq-scale", str(1.0 / shared.args.compress_pos_emb)]
if shared.args.rope_freq_base > 0:
cmd += ["--rope-freq-base", str(shared.args.rope_freq_base)]
# Start the server with pipes for output
self.process = subprocess.Popen(