forked from roger/k3s-cluster
change qwen parallelism to 2
This commit is contained in:
@@ -71,7 +71,7 @@ spec:
|
|||||||
- -c
|
- -c
|
||||||
- "131072"
|
- "131072"
|
||||||
- -np
|
- -np
|
||||||
- "1"
|
- "2"
|
||||||
- --cont-batching
|
- --cont-batching
|
||||||
- --cache-type-k
|
- --cache-type-k
|
||||||
- q8_0
|
- q8_0
|
||||||
|
|||||||
Reference in New Issue
Block a user