forked from roger/k3s-cluster
upgrade qwen 4b context to 65k
This commit is contained in:
@@ -68,7 +68,7 @@ spec:
|
||||
- -ngl
|
||||
- "999"
|
||||
- -c
|
||||
- "8192"
|
||||
- "65536"
|
||||
- -np
|
||||
- "1"
|
||||
- --cont-batching
|
||||
|
||||
Reference in New Issue
Block a user