Spaces:
Sleeping
Sleeping
File size: 2,588 Bytes
51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 51cfed5 59e7020 |
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 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 |
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
# Load model and tokenizer locally
model_name = "GoofyLM/gonzalez-v1"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.float16, # Use float16 for efficiency
device_map="auto" # Automatically distribute across available GPUs/devices
)
def respond(
message,
history: list[tuple[str, str]],
system_message,
max_tokens,
temperature,
top_p,
):
# Format messages for the model
messages = [{"role": "system", "content": system_message}]
for user_msg, assistant_msg in history:
if user_msg:
messages.append({"role": "user", "content": user_msg})
if assistant_msg:
messages.append({"role": "assistant", "content": assistant_msg})
messages.append({"role": "user", "content": message})
# Convert messages to model input format
chat_template = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
# Tokenize the input
inputs = tokenizer(chat_template, return_tensors="pt").to(model.device)
# Generate response with streaming
input_length = inputs.input_ids.shape[1]
generated_tokens = []
# Set up generation parameters
gen_kwargs = {
"max_new_tokens": max_tokens,
"temperature": temperature,
"top_p": top_p,
"do_sample": temperature > 0,
"pad_token_id": tokenizer.eos_token_id,
}
# Stream the generation
response = ""
for output in model.generate(
**inputs,
**gen_kwargs,
streamer=transformers.TextStreamer(tokenizer, skip_prompt=True),
):
# Skip input tokens
if len(output) <= input_length:
continue
# Get new tokens
new_tokens = output[input_length:]
decoded = tokenizer.decode(new_tokens, skip_special_tokens=True)
response = decoded
yield response
demo = gr.ChatInterface(
respond,
additional_inputs=[
gr.Textbox(value="You are a Gonzalez-v1.", 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=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"
),
],
)
if __name__ == "__main__":
demo.launch() |