|
import streamlit as st |
|
from huggingface_hub import InferenceClient |
|
|
|
|
|
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1") |
|
|
|
|
|
def format_prompt(message, history): |
|
prompt = "<s>" |
|
for user_prompt, bot_response in history: |
|
prompt += f"[INST] {user_prompt} [/INST]" |
|
prompt += f" {bot_response}</s> " |
|
prompt += f"[INST] {message} [/INST]" |
|
return prompt |
|
|
|
|
|
def generate(prompt, history, temperature=0.2, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0): |
|
temperature = max(float(temperature), 1e-2) |
|
top_p = float(top_p) |
|
|
|
generate_kwargs = dict( |
|
temperature=temperature, |
|
max_new_tokens=max_new_tokens, |
|
top_p=top_p, |
|
repetition_penalty=repetition_penalty, |
|
do_sample=True, |
|
seed=42, |
|
) |
|
|
|
formatted_prompt = format_prompt(prompt, history) |
|
|
|
stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False) |
|
output = "" |
|
for response in stream: |
|
output += response.token.text |
|
return output |
|
|
|
|
|
st.title("Mistral 8x7b Chat") |
|
|
|
|
|
if 'history' not in st.session_state: |
|
st.session_state.history = [] |
|
|
|
|
|
user_input = st.text_input("Your message:", key="user_input") |
|
|
|
|
|
if st.button("Send"): |
|
if user_input: |
|
bot_response = generate(user_input, st.session_state.history) |
|
st.session_state.history.append((user_input, bot_response)) |
|
|
|
|
|
|
|
chat_text = "" |
|
for user_msg, bot_msg in st.session_state.history: |
|
chat_text += f"You: {user_msg}\nBot: {bot_msg}\n\n" |
|
st.text_area("Chat", value=chat_text, height=300, disabled=False) |
|
|
