add qwen3 no thinking
This commit is contained in:
20
apps/llama/configs/config.yaml
Normal file
20
apps/llama/configs/config.yaml
Normal file
@@ -0,0 +1,20 @@
|
||||
models:
|
||||
"DeepSeek-R1-0528-Qwen3-8B-GGUF":
|
||||
cmd: |
|
||||
/app/llama-server
|
||||
-hf unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF:Q4_K_M
|
||||
-ngl 37 -c 16384
|
||||
--port ${PORT}
|
||||
"Qwen3-8B-GGUF":
|
||||
cmd: |
|
||||
/app/llama-server
|
||||
-hf unsloth/Qwen3-8B-GGUF:Q4_K_M
|
||||
-ngl 37 -c 16384
|
||||
--port ${PORT}
|
||||
"Qwen3-8B-GGUF-no-thinking":
|
||||
cmd: |
|
||||
/app/llama-server
|
||||
-hf unsloth/Qwen3-8B-GGUF:Q4_K_M
|
||||
-ngl 37 -c 16384
|
||||
--jinja --chat-template-file /config/qwen_nothink_chat_template.jinja
|
||||
--port ${PORT}
|
||||
Reference in New Issue
Block a user