take everything that made this app special, and eliminate it

app.py

@@ -1,84 +1,55 @@
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM
-from spaces import GPU as gpu
+import spaces
+from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
+from threading import Thread
+import torch
 
+MODEL_ID = "nroggendorff/smallama-it"
 
-class Delta:
-    def __init__(self, content):
-        self.content = content
-
-
-class Choice:
-    def __init__(self, delta):
-        self.delta = delta
-
-
-class InferenceClient:
-    def __init__(self, model_id="nroggendorff/smallama-it"):
-        self.tokenizer = AutoTokenizer.from_pretrained(model_id)
-        self.model = AutoModelForCausalLM.from_pretrained(model_id)
-
-    class ModelOutput:
-        def __init__(self, client, inputs):
-            self.client = client
-            self.inputs = inputs
-            self.choices = []
-
-        def decode(self, output):
-            decoded_output = self.client.tokenizer.decode(
-                output[0][self.inputs["input_ids"].shape[-1] :],
-                skip_special_tokens=True,
-            )
-            self.choices = [Choice(Delta(decoded_output))]
-            return self
-
-    @gpu
-    def chat_completion(
-        self, messages, max_tokens=256, stream=True, temperature=0.2, top_p=0.95
-    ):
-        inputs = self.tokenizer.apply_chat_template(
-            messages,
-            add_generation_prompt=True,
-            tokenize=True,
-            return_dict=True,
-            return_tensors="pt",
-        ).to(self.model.device)
-
-        model_output = self.ModelOutput(self, inputs)
-
-        for _ in range(max_tokens):
-            output = self.model.generate(
-                **inputs, max_new_tokens=1, temperature=temperature, top_p=top_p
-            )
-            yield model_output.decode(output)
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_ID, dtype=torch.float16, device_map="auto"
+)
 
 
+@spaces.GPU
 def respond(
     message,
     history: list[dict[str, str]],
-    system_message,
     max_tokens,
     temperature,
     top_p,
 ):
-    client = InferenceClient()
-    messages = [{"role": "system", "content": system_message}]
-    messages.extend(history)
+    messages = history
     messages.append({"role": "user", "content": message})
 
-    response = ""
-
-    for message in client.chat_completion(
+    inputs = tokenizer.apply_chat_template(
         messages,
-        max_tokens=max_tokens,
-        stream=True,
+        add_generation_prompt=True,
+        tokenize=True,
+        return_dict=True,
+        return_tensors="pt",
+    ).to(model.device)
+
+    streamer = TextIteratorStreamer(
+        tokenizer, skip_prompt=True, skip_special_tokens=True
+    )
+
+    generation_kwargs = dict(
+        input_ids=inputs["input_ids"],
+        attention_mask=inputs["attention_mask"],
+        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
+        do_sample=True,
+        streamer=streamer,
+    )
+
+    thread = Thread(target=model.generate, kwargs=generation_kwargs)
+    thread.start()
+
+    response = ""
+    for token in streamer:
         response += token
         yield response
 
@@ -88,7 +59,7 @@ chatbot = gr.ChatInterface(
     type="messages",
     additional_inputs=[
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.2, step=0.1, label="Temperature"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,