Update app.py

app.py

@@ -1,54 +1,21 @@
-import asyncio
 import gradio as gr
-from huggingface_hub import AsyncInferenceClient
+from openai import AsyncInferenceClient
 
+# Assuming client is a global variable
 client = AsyncInferenceClient("meta-llama/Llama-2-70b-chat-hf")
 
-system_message = "\nYou are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.\n\nIf a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information."
-
-title = "Llama2 70B Chatbot"
-description = """
-This Space demonstrates model [Llama-2-70b-chat-hf](https://huggingface.co/meta-llama/Llama-2-70b-chat-hf) by Meta, a Llama 2 model with 70B parameters fine-tuned for chat instructions. This space is running on Inference Endpoints using text-generation-inference library. If you want to run your own service, you can also [deploy the model on Inference Endpoints](https://ui.endpoints.huggingface.co/).
-🔎 For more details about the Llama 2 family of models and how to use them with `transformers`, take a look [at our blog post](https://huggingface.co/blog/llama2).
-🔨 Looking for lighter chat model versions of Llama-v2? 
-- 🐇 Check out the [7B Chat model demo](https://huggingface.co/spaces/huggingface-projects/llama-2-7b-chat).
-- 🦊 Check out the [13B Chat model demo](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat).
-Note: As a derivate work of [Llama-2-70b-chat](https://huggingface.co/meta-llama/Llama-2-70b-chat-hf) by Meta,
-this demo is governed by the original [license](https://huggingface.co/spaces/ysharma/Explore_llamav2_with_TGI/blob/main/LICENSE.txt) and [acceptable use policy](https://huggingface.co/spaces/ysharma/Explore_llamav2_with_TGI/blob/main/USE_POLICY.md).
-"""
-css = """.toast-wrap { display: none !important } """
-examples = [
-    ['Hello there! How are you doing?'],
-    ['Can you explain to me briefly what is Python programming language?'],
-    ['Explain the plot of Cinderella in a sentence.'],
-    ['How many hours does it take a man to eat a Helicopter?'],
-    ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
-]
-
-# Note: We have removed default system prompt as requested by the paper authors [Dated: 13/Oct/2023]
-# Prompting style for Llama2 without using system prompt
-# <s>[INST] {{ user_msg_1 }} [/INST] {{ model_answer_1 }} </s><s>[INST] {{ user_msg_2 }} [/INST]
-
-# Stream text - stream tokens with InferenceClient from TGI
-async def predict(message, chatbot, system_prompt="", temperature=0.9, max_new_tokens=256, top_p=0.6, repetition_penalty=1.0,):
-
-    if system_prompt != "":
-        input_prompt = f"<s>[INST] <<SYS>>\n{system_prompt}\n<</SYS>>\n\n "
-    else:
-        input_prompt = f"<s>[INST] "
-
-    temperature = float(temperature)
-    if temperature < 1e-2:
-        temperature = 1e-2
+def predict(message, chatbot, system_prompt="", temperature=0.9, max_new_tokens=256, top_p=0.6, repetition_penalty=1.0):
+    input_prompt = f"<s>[INST] <<SYS>>\n{system_prompt}\n<</SYS>>\n\n " if system_prompt else "<s>[INST] "
+    temperature = max(1e-2, float(temperature))
     top_p = float(top_p)
 
     for interaction in chatbot:
-        input_prompt = input_prompt + str(interaction[0]) + " [/INST] " + str(interaction[1]) + " </s><s>[INST] "
-
-    input_prompt = input_prompt + str(message) + " [/INST] "
+        input_prompt += f"{interaction[0]} [/INST] {interaction[1]} </s><s>[INST] "
 
+    input_prompt += f"{message} [/INST] "
     partial_message = ""
-    async for token in await client.text_generation(
+
+    for token in client.text_generation(
         prompt=input_prompt,
         max_new_tokens=max_new_tokens,
         stream=True,
@@ -58,140 +25,22 @@ async def predict(message, chatbot, system_prompt="", temperature=0.9, max_new_t
         do_sample=True,
         repetition_penalty=repetition_penalty,
     ):
-        partial_message = partial_message + token
+        partial_message += token
         yield partial_message
 
-
-# No Stream - batch produce tokens using TGI inference endpoint
-def predict_batch(
-    message, chatbot, system_prompt="", temperature=0.9, max_new_tokens=256, top_p=0.6, repetition_penalty=1.0,
-):
-
-    if system_prompt != "":
-        input_prompt = f"<s>[INST] <<SYS>>\n{system_prompt}\n<</SYS>>\n\n "
-    else:
-        input_prompt = f"<s>[INST] "
-
-    temperature = float(temperature)
-    if temperature < 1e-2:
-        temperature = 1e-2
-    top_p = float(top_p)
-
-    for interaction in chatbot:
-        input_prompt = input_prompt + str(interaction[0]) + " [/INST] " + str(interaction[1]) + " </s><s>[INST] "
-
-    input_prompt = input_prompt + str(message) + " [/INST] "
-    print(f"input_prompt - {input_prompt}")
-
-    data = {
-        "inputs": input_prompt,
-        "parameters": {
-            "max_new_tokens": max_new_tokens,
-            "temperature": temperature,
-            "top_p": top_p,
-            "repetition_penalty": repetition_penalty,
-            "do_sample": True,
-        },
-    }
-
-    response = requests.post(api_url, headers=headers, json=data)  # auth=('hf', hf_token)) data=json.dumps(data),
-
-    if response.status_code == 200:  # check if the request was successful
-        try:
-            json_obj = response.json()
-            if 'generated_text' in json_obj[0] and len(json_obj[0]['generated_text']) > 0:
-                return json_obj[0]['generated_text']
-            elif 'error' in json_obj[0]:
-                return json_obj[0]['error'] + ' Please refresh and try again with smaller input prompt'
-            else:
-                print(f"Unexpected response: {json_obj[0]}")
-        except json.JSONDecodeError:
-            print(f"Failed to decode response as JSON: {response.text}")
-    else:
-        print(f"Request failed with status code {response.status_code}")
-
-
-def vote(data: gr.LikeData):
-    if data.liked:
-        print("You upvoted this response: " + data.value)
-    else:
-        print("You downvoted this response: " + data.value)
-
-
-additional_inputs = [
-    gr.Textbox("", label="Optional system prompt"),
-    gr.Slider(
-        label="Temperature",
-        value=0.9,
-        minimum=0.0,
-        maximum=1.0,
-        step=0.05,
-        interactive=True,
-        info="Higher values produce more diverse outputs",
-    ),
-    gr.Slider(
-        label="Max new tokens",
-        value=256,
-        minimum=0,
-        maximum=4096,
-        step=64,
-        interactive=True,
-        info="The maximum numbers of new tokens",
-    ),
-    gr.Slider(
-        label="Top-p (nucleus sampling)",
-        value=0.6,
-        minimum=0.0,
-        maximum=1,
-        step=0.05,
-        interactive=True,
-        info="Higher values sample more low-probability tokens",
-    ),
-    gr.Slider(
-        label="Repetition penalty",
-        value=1.2,
-        minimum=1.0,
-        maximum=2.0,
-        step=0.05,
-        interactive=True,
-        info="Penalize repeated tokens",
-    )
-]
-
-chatbot_stream = gr.Chatbot(bubble_full_width=False)
-chatbot_batch = gr.Chatbot(bubble_full_width=False)
-chat_interface_stream = gr.ChatInterface(
-    predict,
-    title=title,
-    description=description,
-    textbox=gr.Textbox(),
-    chatbot=chatbot_stream,
-    css=css,
-    examples=examples,
-    additional_inputs=additional_inputs,
-)
-chat_interface_batch = gr.ChatInterface(
-    predict_batch,
-    title=title,
-    description=description,
-    textbox=gr.Textbox(),
-    chatbot=chatbot_batch,
-    css=css,
-    examples=examples,
-    additional_inputs=additional_inputs,
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=predict,
+    inputs=[
+        gr.Textbox("text", label="Message"),
+        gr.Textbox("text", label="Chatbot"),
+        gr.Textbox("text", label="System Prompt"),
+        gr.Number("slider", minimum=0.1, maximum=2, default=0.9, label="Temperature"),
+        gr.Number("slider", minimum=1, maximum=1000, default=256, label="Max New Tokens"),
+        gr.Number("slider", minimum=0.1, maximum=1, default=0.6, label="Top P"),
+        gr.Number("slider", minimum=0.1, maximum=2, default=1.0, label="Repetition Penalty"),
+    ],
+    outputs=gr.Textbox(),
 )
 
-# Gradio Demo
-with gr.Blocks() as demo:
-
-    with gr.Tab("Streaming"):
-        # streaming chatbot
-        chatbot_stream.like(vote, None, None)
-        chat_interface_stream.render()
-
-    with gr.Tab("Batch"):
-        # non-streaming chatbot
-        chatbot_batch.like(vote, None, None)
-        chat_interface_batch.render()
-
-demo.queue(max_size=100).launch()
+iface.launch()