new stuff

app.py

@@ -1,61 +1,52 @@
+import os
 import gradio as gr
 from huggingface_hub import InferenceClient
 
 ENDPOINT_URL = "https://x6leavj4hgm2fdyx.us-east-2.aws.endpoints.huggingface.cloud"
 
-def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
-):
-    """
-    Uses your Hugging Face Inference Endpoint for chat completion.
-    """
-    # Use the endpoint URL here. `token` must be the raw string.
-    client = InferenceClient(model=ENDPOINT_URL, token=hf_token.token)
-
-    # Build messages (system + prior turns + current user)
-    messages = [{"role": "system", "content": system_message}]
-    messages.extend(history)
-    messages.append({"role": "user", "content": message})
-
-    # Stream tokens from the endpoint
-    response = ""
-    for chunk in client.chat_completion(
-        messages=messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = ""
-        if chunk.choices and chunk.choices[0].delta.content:
-            token = chunk.choices[0].delta.content
-        response += token
-        yield response
-
-
-# ---- Gradio UI ----
+def respond(message, history, system_message, max_tokens, temperature, top_p, hf_pat):
+    token = hf_pat or os.getenv("HUGGINGFACEHUB_API_TOKEN")
+    if not token:
+        yield "Missing token. Paste a Hugging Face PAT with endpoint write access."
+        return
+
+    client = InferenceClient(model=ENDPOINT_URL, token=token)
+
+    messages = [{"role": "system", "content": system_message}, *history,
+                {"role": "user", "content": message}]
+
+    try:
+        out = ""
+        for chunk in client.chat_completion(
+            messages=messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            delta = chunk.choices[0].delta.content if chunk.choices and chunk.choices[0].delta else ""
+            if delta:
+                out += delta
+                yield out
+    except Exception as e:
+        yield f"Request failed: {e}\n\nCheck that your token has `inference.endpoints.infer.write` and access to this endpoint."
+
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
     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.Slider(1, 2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(0.1, 4.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
+        gr.Textbox(lines=1, type="password", placeholder="hf_... token", label="HF PAT (with endpoint write)"),
     ],
 )
 
 with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()  # lets you pass the HF token to the app
     chatbot.render()
 
 if __name__ == "__main__":
+    # Or export HUGGINGFACEHUB_API_TOKEN in your shell and leave the textbox empty.
     demo.launch()