No GPU - Q8

app.py

@@ -8,7 +8,7 @@ For more information on `huggingface_hub` Inference API support, please check th
 """
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-pipe = pipeline("text-generation", model="HiTZ/latxa-7b-v1", device=0)
+pipe = pipeline("text-generation", model="HiTZ/latxa-7b-v1", device_map="auto", model_kwargs={"load_in_8bit": True})
 
 
 def respond(

requirements.txt

@@ -1,3 +1,5 @@
 huggingface_hub==0.22.2
 transformers
-torch
+torch
+accelerate 
+bitsandbytes