Parameters: change max_new_tokens & repetition_penalty_range defaults (#4842)

extensions/openai/completions.py

@@ -236,7 +236,7 @@ def chat_completions_common(body: dict, is_legacy: bool = False, stream=False) -
 
     max_tokens = generate_params['max_new_tokens']
     if max_tokens in [None, 0]:
-        generate_params['max_new_tokens'] = 200
+        generate_params['max_new_tokens'] = 512
         generate_params['auto_max_new_tokens'] = True
 
     requested_model = generate_params.pop('model')

extensions/openai/typing.py

@@ -10,7 +10,7 @@ class GenerationOptions(BaseModel):
     min_p: float = 0
     top_k: int = 0
     repetition_penalty: float = 1
-    repetition_penalty_range: int = 0
+    repetition_penalty_range: int = 1024
     typical_p: float = 1
     tfs: float = 1
     top_a: float = 0