Update app.py

app.py

@@ -27,10 +27,10 @@ cache = Cache()
 hf_token = os.environ.get("HF_TOKEN")
 
 llm_models = {
-    "4": VLLM(model="lilmeaty/4", trust_remote_code=True, max_new_tokens=50, temperature=0.1, use_auth_token=hf_token, device="cpu"),
-    "yi-coder": VLLM(model="01-ai/Yi-Coder-1.5B", trust_remote_code=True, max_new_tokens=50, temperature=0.6, use_auth_token=hf_token, device="cpu"),
-    "llama": VLLM(model="meta-llama/Llama-3.2-3B-Instruct", trust_remote_code=True, max_new_tokens=50, temperature=0.1, use_auth_token=hf_token, device="cpu"),
-    "qwen": VLLM(model="Qwen/Qwen2.5-1.5B-Instruct", trust_remote_code=True, max_new_tokens=50, temperature=0.6, use_auth_token=hf_token, device="cpu"),
+    "4": VLLM(model="lilmeaty/4", trust_remote_code=True, use_cuda=False, max_new_tokens=50, temperature=0.1, use_auth_token=hf_token, device="cpu"),
+    "yi-coder": VLLM(model="01-ai/Yi-Coder-1.5B", trust_remote_code=True, use_cuda=False, max_new_tokens=50, temperature=0.6, use_auth_token=hf_token, device="cpu"),
+    "llama": VLLM(model="meta-llama/Llama-3.2-3B-Instruct", trust_remote_code=True, use_cuda=False, max_new_tokens=50, temperature=0.1, use_auth_token=hf_token, device="cpu"),
+    "qwen": VLLM(model="Qwen/Qwen2.5-1.5B-Instruct", trust_remote_code=True, use_cuda=False, max_new_tokens=50, temperature=0.6, use_auth_token=hf_token, device="cpu"),
 }
 
 for llm_name, llm in llm_models.items():