removed huggingface_hub

backend.py

@@ -13,10 +13,10 @@ from llama_cpp import Llama
 import spaces
 
 
-huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
+#huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
 
 
-hf_hub_download(
+"""hf_hub_download(
     repo_id="google/gemma-2-2b-it-GGUF",
     filename="2b_it_v2.gguf",
     local_dir="./models",
@@ -29,20 +29,20 @@ llm = Llama(
             _gpu_layers=81,
             n_batch=1024,
             n_ctx=8192,
-        )
+        )"""
 
 
-"""model_id = "google/gemma-2-2b-it"
+model_id = "google/gemma-2-2b-it"
 tokenizer = GemmaTokenizerFast.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
     device_map="auto",
     torch_dtype= torch.float16 if torch.cuda.is_available() else torch.float32,
-)"""
+)
 # what models will be used by LlamaIndex:
 Settings.embed_model = InstructorEmbedding(model_name="hkunlp/instructor-base")
-#Settings.llm  = GemmaLLMInterface(model=model, tokenizer=tokenizer)
-Settings.llm  = llm
+Settings.llm  = GemmaLLMInterface(model=model, tokenizer=tokenizer)
+#Settings.llm  = llm
 
 
 ############################---------------------------------