test

app.py

@@ -4,13 +4,17 @@ import torch
 import subprocess
 import numpy as np
 
-zero = torch.Tensor([0]).cuda()
-print(zero.device) # <-- 'cpu' 🤔
-
+device_available = print(torch.cuda.is_available())  # <-- True 🤗
+device_count = print(torch.cuda.device_count())  # <-- 1 🤗
+device_name = print(torch.cuda.get_device_name())  # <-- 'A10G' 🤗
 @spaces.GPU
 def start_ochat_server():
-    print(zero.device)  # <-- 'cuda:0' 🤗
+    global device_available, device_count, device_name
 
+    print(device_available)
+    print(device_name)  # <-- 'cuda:0' 🤗
+    print(device_count)  # <-- 1 🤗
+    
     # Command to start the ochat inference server
     command = [
         "python", "-m", "ochat.serving.openai_api_server", 

requirements.txt

@@ -1,7 +1,7 @@
 git+https://github.com/huggingface/transformers.git
-git+https://github.com/vllm-project/vllm.git
 --extra-index-url https://download.pytorch.org/whl/cu113
 torch
+git+https://github.com/vllm-project/vllm.git
 datasets
 accelerate
 numpy