File size: 1,770 Bytes
2b6b154 d24a963 1b3204d 145ecb9 d24a963 145ecb9 2b6b154 4581f8a 145ecb9 4581f8a 145ecb9 afd19d1 145ecb9 4581f8a 145ecb9 afd19d1 145ecb9 2b6b154 09e1b8b 145ecb9 afd19d1 1b3204d 4581f8a afd19d1 4581f8a 09e1b8b 4581f8a 09e1b8b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
import gradio as gr
import spaces
import torch
import transformers
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "meta-llama/Meta-Llama-3-8B-Instruct"
pipeline = transformers.pipeline(
"text-generation",
model=model_name,
model_kwargs={"torch_dtype": torch.bfloat16},
device="cuda",
)
@spaces.GPU
def chat_function(message, history, system_prompt,max_new_tokens,temperature):
messages = [
{"role": "system", "content": system_prompt},
{"role": "user", "content": message},
]
prompt = pipeline.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
terminators = [
pipeline.tokenizer.eos_token_id,
pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]
temp = temperature + 0.1
outputs = pipeline(
prompt,
max_new_tokens=max_new_tokens,
eos_token_id=terminators,
do_sample=True,
temperature=temp,
top_p=0.9,
)
return outputs[0]["generated_text"][len(prompt):]
gr.ChatInterface(
chat_function,
chatbot=gr.Chatbot(height=500),
textbox=gr.Textbox(placeholder="Enter message here", container=False, scale=7),
title="LLAMA 3 8B Chat",
description="""
This space is dedicated for chatting with Meta Latest LLM - Llama 8b Instruct. Find this model here: https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct
Feel free to play with customization in the "Additional Inputs".
""",
theme="soft",
additional_inputs=[
gr.Textbox("You are helpful AI.", label="System Prompt"),
gr.Slider(512, 4096, label="Max New Tokens"),
gr.Slider(0, 1, label="Temperature")
]
).launch() |