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