...

app.py

@@ -1,7 +1,8 @@
 import streamlit as st
-from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
+from transformers import AutoTokenizer
 import time
-import requests
+import asyncio
+import httpx
 import json
 
 repo_name = "BeardedMonster/SabiYarn-125M"
@@ -62,18 +63,19 @@ st.write("**Model is running on CPU RAM. So, token generation might be slower (s
 st.write("**Avg Response time: 15 secs/ 50 tokens. Response time increase with input length**")
 st.write("-" * 50)
 
-def generate_from_api(user_input, generation_config):
-    url = " https://pauljeffrey--sabiyarn-fastapi-app.modal.run/predict"
-
+async def generate_from_api(user_input, generation_config):
+    url = "https://pauljeffrey--sabiyarn-fastapi-app.modal.run/predict"
+    
     payload = json.dumps({
-      "prompt": user_input,
-      "config": generation_config
+        "prompt": user_input,
+        "config": generation_config
     })
     headers = {
-      'Content-Type': 'application/json'
+        'Content-Type': 'application/json'
     }
     
-    response = requests.request("POST", url, headers=headers, data=payload)
+    async with httpx.AsyncClient() as client:
+        response = await client.post(url, headers=headers, data=payload)
     return response.json()["generated_text"]
 
 # Sample texts
@@ -108,7 +110,7 @@ if st.button("Generate"):
         try:
             st.write("**Generated Text Below:**")
             wrapped_input = task_options[task].format(user_input)
-            generated_text = generate_from_api(wrapped_input, generation_config)
+            generated_text = asyncio.run(generate_from_api(wrapped_input, generation_config))
             full_output = st.empty()
            
             start_time = time.time()