Update app.py

app.py

@@ -1,61 +1,74 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+from threading import Thread
 
+model_id = "sapientinc/HRM-Text-1B"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    torch_dtype=torch.bfloat16,
+    trust_remote_code=True,
+    device_map="auto"
+).eval()
+
+condition = "<|quad_end|><|object_ref_end|>"
+
+def build_prompt(system_message: str, history: list[dict[str, str]], user_message: str) -> str:
+    prompt = condition
+    if system_message:
+        prompt += f"<|im_start|>system\n{system_message}<|im_end|>"
+    
+    for turn in history:
+        if turn["role"] == "user":
+            prompt += f"{condition}<|im_start|>user\n{turn['content']}<|im_end|>"
+        elif turn["role"] == "assistant":
+            prompt += f"<|im_start|>assistant\n{turn['content']}<|im_end|>"
+    
+    prompt += f"{condition}<|im_start|>user\n{user_message}<|im_end|><|im_start|>assistant\n"
+    return prompt
 
 def respond(
-    message,
+    message: str,
     history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
+    system_message: str,
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
+    hf_token: gr.OAuthToken = None,
 ):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
-
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    prompt = build_prompt(system_message, history, message)
+    
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    inputs["token_type_ids"] = torch.ones_like(inputs["input_ids"])
+    
+    streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=False)
+    
+    generation_kwargs = dict(
+        **inputs,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-
-        response += token
-        yield response
+        do_sample=temperature > 0,
+        streamer=streamer,
+        pad_token_id=tokenizer.eos_token_id,
+    )
+    
+    thread = Thread(target=model.generate, kwargs=generation_kwargs)
+    thread.start()
+    
+    generated_text = ""
+    for new_token in streamer:
+        generated_text += new_token
+        yield generated_text
 
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
 chatbot = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Textbox(value="You are a helpful reasoning assistant.", 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(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
     ],
 )
 
@@ -64,6 +77,5 @@ with gr.Blocks() as demo:
         gr.LoginButton()
     chatbot.render()
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()