Update app.py

app.py

@@ -12,8 +12,8 @@ print(f"Is CUDA available: {torch.cuda.is_available()}")
 print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 
 pipe_biogpt = pipeline("text-generation", model="microsoft/biogpt")
-pipe_biomedlm = pipeline("text-generation", model="stanford-crfm/BioMedLM", device="cuda:0")
-pipe_galactica = pipeline("text-generation", model="facebook/galactica-1.3b", device="cuda:0")
+pipe_biomedlm = pipeline("text-generation", model="stanford-crfm/BioMedLM", device="cuda:0", model_kwargs={"torch_dtype":torch.bfloat16})
+pipe_galactica = pipeline("text-generation", model="facebook/galactica-1.3b", device="cuda:0", model_kwargs={"torch_dtype":torch.bfloat16})
 
 title = "Compare generative biomedical LLMs!"
 description = "**Disclaimer:** this demo was made for research purposes only and should not be used for medical purposes."