fix

app.py

@@ -1,18 +1,14 @@
 import gradio as gr
 import os
-import cohere
 from huggingface_hub import InferenceClient
+import cohere
 
-# Retrieve API keys from environment variables
-HF_API_KEY = os.getenv("HF_API_KEY")  
-COHERE_API_KEY = os.getenv("COHERE_API_KEY")  
-
-# Initialize clients
-hf_model = "meta-llama/Llama-3.2-3B-Instruct"  # Change to preferred HF model
-hf_client = InferenceClient(model=hf_model, token=HF_API_KEY)
-
-cohere_client = cohere.Client(COHERE_API_KEY)
+HF_API_KEY = os.getenv("HF_API_KEY")
+COHERE_API_KEY = os.getenv("COHERE_API_KEY")  # Get Cohere API key
 
+models = ["HuggingFaceH4/zephyr-7b-beta", "microsoft/Phi-4-mini-instruct", "meta-llama/Llama-3.2-3B-Instruct"]
+client_hf = InferenceClient(model=models[2], token=HF_API_KEY)  # HF Client
+client_cohere = cohere.Client(COHERE_API_KEY)  # Cohere Client
 
 def respond(
     message,
@@ -21,7 +17,7 @@ def respond(
     max_tokens,
     temperature,
     top_p,
-    use_cohere,  # Checkbox input
+    use_cohere,  # Checkbox value
 ):
     messages = [{"role": "system", "content": system_message}]
     
@@ -32,23 +28,21 @@ def respond(
             messages.append({"role": "assistant", "content": val[1]})
 
     messages.append({"role": "user", "content": message})
+    
+    response = ""
 
-    # 🔹 **Switch API based on checkbox**
-    if use_cohere:
-        response = cohere_client.chat(
-            model="command-r-plus",  
+    if use_cohere:  # If Cohere is selected
+        cohere_response = client_cohere.chat(
             message=message,
-            chat_history=[{"user_name": "User", "text": h[0]} for h in history if h[0]] +
-                         [{"user_name": "Assistant", "text": h[1]} for h in history if h[1]],
+            model="command-r",  # Or "command" depending on your plan
             temperature=temperature,
-            max_tokens=max_tokens,
-            p=top_p,
+            max_tokens=max_tokens
         )
-        return response.text  # Cohere returns full response
-
-    else:
-        response = ""
-        for message in hf_client.chat_completion(
+        response = cohere_response.text
+        yield response  # Yield full response (Cohere doesn't stream)
+    
+    else:  # If HF is selected
+        for message in client_hf.chat_completion(
             messages,
             max_tokens=max_tokens,
             stream=True,
@@ -57,24 +51,17 @@ def respond(
         ):
             token = message.choices[0].delta.content
             response += token
-            yield response  # Hugging Face supports streaming
-
+            yield response
 
-# 🔥 **Gradio UI with Checkbox**
+# Gradio UI
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-        gr.Checkbox(label="Use Cohere API", value=False),  # Checkbox to toggle API
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
+        gr.Checkbox(label="Use Cohere API"),  # Checkbox to switch API
     ],
 )