mock responses for streaming

litellm/main.py

@@ -954,6 +954,30 @@ def batch_completion(
     return results
 
 
+def mock_completion(model: str, messages: List, stream: bool = False, mock_response: str = "This is a mock request"):
+    try:
+        model_response = ModelResponse()
+        if stream: # return a generator object, iterate through the text in chunks of 3 char / chunk
+            for i in range(0, len(mock_response), 3):
+                completion_obj = {"role": "assistant", "content": mock_response[i: i+3]}
+                yield {
+                "choices": 
+                    [
+                        {
+                            "delta": completion_obj,
+                            "finish_reason": None
+                        },
+                    ]
+                }
+        else:
+            ## RESPONSE OBJECT
+            completion_response = "This is a mock request"
+            model_response["choices"][0]["message"]["content"] = completion_response
+            model_response["created"] = time.time()
+            model_response["model"] = "MockResponse"
+            return model_response
+    except:
+        raise Exception("Mock completion response failed")
 ### EMBEDDING ENDPOINTS ####################
 @client
 @timeout(  # type: ignore