Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,3 +1,32 @@
|
|
1 |
import gradio as gr
|
|
|
|
|
2 |
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
3 |
+
import torch
|
4 |
|
5 |
+
# Load the model and tokenizer
|
6 |
+
model_name = "mattshumer/Reflection-Llama-3.1-70B"
|
7 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
8 |
+
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
|
9 |
+
|
10 |
+
def generate_response(message, history):
|
11 |
+
# Combine history and new message
|
12 |
+
prompt = "\n".join([f"Human: {h[0]}\nAI: {h[1]}" for h in history])
|
13 |
+
prompt += f"\nHuman: {message}\nAI:"
|
14 |
+
|
15 |
+
# Tokenize and generate
|
16 |
+
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
|
17 |
+
outputs = model.generate(**inputs, max_new_tokens=500, temperature=0.7)
|
18 |
+
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
19 |
+
|
20 |
+
# Extract only the AI's response
|
21 |
+
ai_response = response.split("AI:")[-1].strip()
|
22 |
+
return ai_response
|
23 |
+
|
24 |
+
# Create the Gradio interface
|
25 |
+
iface = gr.ChatInterface(
|
26 |
+
fn=generate_response,
|
27 |
+
title="Chat with Reflection-Llama-3.1-70B",
|
28 |
+
description="Ask me anything!",
|
29 |
+
)
|
30 |
+
|
31 |
+
# Launch the interface
|
32 |
+
iface.launch()
|