Spaces:
Sleeping
Sleeping
File size: 2,680 Bytes
171f2ef 8fca8cb d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef d1912a0 171f2ef |
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 85 86 87 88 |
import gradio as gr
import spaces
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
# 只加载一次模型和分词器
MODEL_NAME = "inclusionAI/Ring-mini-2.0"
device = "cuda" if torch.cuda.is_available() else "cpu"
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
MODEL_NAME,
torch_dtype=torch.float16 if device == "cuda" else torch.float32,
trust_remote_code=True
).to(device)
@spaces.GPU
def respond(
message,
history: list[dict[str, str]],
system_message,
max_tokens,
temperature,
top_p,
hf_token: gr.OAuthToken = None, # 保持参数兼容
):
"""
使用 transformers 在 GPU 上本地推理 inclusionAI/Ring-mini-2.0
"""
# 拼接历史和 system prompt，兼容 gradio ChatInterface 的消息格式
prompt = system_message + "\n"
# gradio history: [{"role": "user"/"assistant", "content": "..."}, ...]
last_role = None
for turn in history:
if turn.get("role") == "user":
prompt += f"User: {turn['content']}\n"
last_role = "user"
elif turn.get("role") == "assistant":
prompt += f"Assistant: {turn['content']}\n"
last_role = "assistant"
prompt += f"User: {message}\nAssistant:"
input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(device)
output_ids = model.generate(
input_ids,
max_new_tokens=max_tokens,
temperature=temperature,
top_p=top_p,
do_sample=True,
pad_token_id=tokenizer.eos_token_id,
)
output = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
# 流式输出
response = ""
for token in output.split():
response += token + " "
yield response.strip()
"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
chatbot = gr.ChatInterface(
respond,
type="messages",
additional_inputs=[
gr.Textbox(value="You are 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=1.0,
value=0.95,
step=0.05,
label="Top-p (nucleus sampling)",
),
],
)
with gr.Blocks() as demo:
gr.Markdown("# HuggingFace Running")
with gr.Sidebar():
gr.LoginButton()
chatbot.render()
if __name__ == "__main__":
demo.launch()
|