update model info

README.md

@@ -8,6 +8,6 @@ sdk_version: 4.36.1
 app_file: app.py
 pinned: false
 license: llama3
----
-
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+models:
+    - Magpie-Align/Llama-3-8B-Magpie-Align-v0.1
+---

app.py

@@ -1,19 +1,27 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import spaces
 
-"""
-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("HuggingFaceH4/zephyr-7b-beta")
+# Load model and tokenizer
+model_name = "Magpie-Align/Llama-3-8B-Magpie-Align-v0.1"
 
+device = "cuda" # the device to load the model onto
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto"
+)
+model.to(device)
 
+@spaces.GPU(enable_queue=True)
 def respond(
     message,
     history: list[tuple[str, str]],
     system_message,
-    max_tokens,
-    temperature,
-    top_p,
+    max_tokens=2048,
+    temperature=0.6,
+    top_p=0.9,
+    repetition_penalty=1.0,
 ):
     messages = [{"role": "system", "content": system_message}]
 
@@ -24,20 +32,27 @@ def respond(
             messages.append({"role": "assistant", "content": val[1]})
 
     messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
+ 
+    text = tokenizer.apply_chat_template(
         messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        tokenize=False,
+        add_generation_prompt=True
+    )
+    model_inputs = tokenizer([text], return_tensors="pt").to(device)
+
+    generated_ids = model.generate(
+        model_inputs.input_ids,
+        max_new_tokens = max_tokens,
+        temperature = temperature,
+        top_p = top_p,
+        repetition_penalty=repetition_penalty,
+    )
+    generated_ids = [
+        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+    ]
 
-        response += token
-        yield response
+    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    return response
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
@@ -45,19 +60,20 @@ For information on how to customize the ChatInterface, peruse the gradio docs: h
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Textbox(value="You are Magpie, 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=4.0, value=0.6, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,
-            value=0.95,
+            value=0.9,
             step=0.05,
             label="Top-p (nucleus sampling)",
         ),
+        gr.Slider(minimum=0.5, maximum=1.5, value=1.0, step=0.1, label="Repetation Penalty"),
     ],
 )
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(share=True)