Update crew.py

crew.py

@@ -12,12 +12,15 @@ from crewai_tools import (
 from openai import OpenAI
 from openinference.instrumentation.crewai import CrewAIInstrumentor
 from phoenix.otel import register
-from util import get_final_answer
+from util import get_final_answer, get_img_b64
+
+## LLMs
 
 MANAGER_MODEL      = "gpt-4.1"
 AGENT_MODEL        = "gpt-4.1"
 FINAL_ANSWER_MODEL = "gpt-4.5-preview"
-STT_MODEL          = "whisper-1"
+AUDIO_MODEL        = "gpt-4o-transcribe"
+IMAGE_MODEL        = "gpt-4.1"
 
 # LLM evaluation
 
@@ -58,7 +61,7 @@ def run_crew(question, file_path):
 
             transcript = client.audio.transcriptions.create(
                 file=open(file_path, "rb"),
-                model=STT_MODEL,
+                model=AUDIO_MODEL,
                 prompt=question
             )
             
@@ -85,14 +88,19 @@ def run_crew(question, file_path):
             
         try:
             client = OpenAI()
-
-            transcript = client.audio.transcriptions.create(
-                file=open(file_path, "rb"),
-                model=STT_MODEL,
-                prompt=question
+    
+            img_b64 = get_img_b64(file_path)
+    
+            completion = client.chat.completions.create(
+                messages=[{"role": "user",
+                            "content": [{"type": "text", 
+                                         "text": question},
+                                        {"type": "image_url",
+                                         "image_url": {"url": f"data:image/jpeg;base64,{img_b64}"}}]}],
+                model=IMAGE_MODEL
             )
-            
-            return transcript.text
+        
+            return completion.choices[0].message.content
         except Exception as e:
             raise RuntimeError(f"Failed to process image: {str(e)}")