Update app.py

app.py

@@ -1,83 +1,49 @@
 import os
-import gradio as gr
 from openai import OpenAI
+import gradio as gr
 
-# -------------------------------------------------
-# 🔐 1. Secure API setup
-# -------------------------------------------------
-# IMPORTANT:
-# Do NOT hardcode your token.
-# Instead, go to your Space → Settings → Repository secrets
-# Add a secret with:
-#   Key: HF_TOKEN
-#   Value: your Hugging Face token (e.g. hf_xxx...)
-# Then this will safely load it at runtime:
+# --- Setup ---
+# Hugging Face Space will automatically inject your token from Secrets
+HF_TOKEN = os.getenv("HF_TOKEN")
+
+# Create client for Hugging Face Inference Router
 client = OpenAI(
     base_url="https://router.huggingface.co/v1",
-    api_key=os.getenv("HF_TOKEN"),
+    api_key=HF_TOKEN,
 )
 
-# -------------------------------------------------
-# 🧩 2. Chat function (with streaming)
-# -------------------------------------------------
-def chat_with_model(message, history, model_name):
-    messages = []
-    for human, assistant in history:
+# --- Chat Function ---
+def chat_with_model(message, history):
+    # Convert history into OpenAI-compatible messages
+    messages = [{"role": "system", "content": "You are a helpful assistant."}]
+    for human, ai in history:
         messages.append({"role": "user", "content": human})
-        if assistant:
-            messages.append({"role": "assistant", "content": assistant})
+        messages.append({"role": "assistant", "content": ai})
     messages.append({"role": "user", "content": message})
 
-    full_reply = ""
-    stream = client.chat.completions.create(
-        model=model_name,
-        messages=messages,
-        stream=True,
-    )
-
-    for chunk in stream:
-        delta = chunk.choices[0].delta.content
-        if delta:
-            full_reply += delta
-            yield full_reply  # stream live output
-
-# -------------------------------------------------
-# 🎨 3. Gradio interface
-# -------------------------------------------------
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown(
-        """
-        <h2 style="text-align:center;">🤖 GPT-OSS Chat (via Hugging Face Router)</h2>
-        <p style="text-align:center;">
-        Stream responses from Fireworks-AI's GPT-OSS models — 7B, 20B, and 120B — through the Hugging Face Inference Router.
-        </p>
-        """,
-        elem_id="title",
-    )
-
-    with gr.Row():
-        model_selector = gr.Dropdown(
-            label="Select Model",
-            choices=[
-                "openai/gpt-oss-7b:fireworks-ai",
-                "openai/gpt-oss-20b:fireworks-ai",
-                "openai/gpt-oss-120b:fireworks-ai",
-            ],
-            value="openai/gpt-oss-120b:fireworks-ai",
-            interactive=True,
+    try:
+        response = client.chat.completions.create(
+            model="openai/gpt-oss-120b:fireworks-ai",
+            messages=messages,
+            max_tokens=500,
+            temperature=0.7,
         )
+        reply = response.choices[0].message.content
+    except Exception as e:
+        reply = f"⚠️ Error: {str(e)}"
+
+    return reply
+
+# --- Gradio UI ---
+iface = gr.ChatInterface(
+    fn=chat_with_model,
+    chatbot=gr.Chatbot(height=500, type="messages", show_label=False),
+    title="💬 GPT-OSS 120B Chat",
+    description="Chat with a massive open-source language model hosted via Hugging Face Router.",
+    theme="soft",
+    textbox=gr.Textbox(placeholder="Ask me anything...", scale=7),
+)
 
-    gr.ChatInterface(
-        fn=lambda msg, hist: chat_with_model(msg, hist, model_selector.value),
-        title="GPT-OSS Chat",
-        chatbot=gr.Chatbot(height=500, show_label=False),
-        textbox=gr.Textbox(placeholder="Ask me anything...", autofocus=True),
-        retry_btn="↩ Retry",
-        clear_btn="🧹 Clear Chat",
-    )
-
-# -------------------------------------------------
-# 🚀 4. Launch app
-# -------------------------------------------------
+# --- Launch ---
 if __name__ == "__main__":
-    demo.launch()
+    iface.launch()