phase 17: disable qwen3.5 thinking mode (think:false) to reduce latency and improve structured output
This commit is contained in:
@@ -220,6 +220,7 @@ class OllamaClient:
|
||||
],
|
||||
"format": json_schema,
|
||||
"stream": False,
|
||||
"think": False,
|
||||
}
|
||||
|
||||
url = f"{self._config.base_url}/api/chat"
|
||||
|
||||
Reference in New Issue
Block a user