Update app.py

agent.py

@@ -55,6 +55,8 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 from huggingface_hub import login
 from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline, BitsAndBytesConfig
 
+from langchain_huggingface import HuggingFaceEndpoint
+
 load_dotenv()
 
 @tool
@@ -459,47 +461,17 @@ tools = [wiki_tool, calc_tool, file_tool, web_tool, arvix_tool, youtube_tool, vi
 
 
 # Get the Hugging Face API token from the environment variable
-hf_token = os.getenv("HF_TOKEN")
-
-# Load Hugging Face token
-login(token=hf_token)
-
-# Model name
-model_name = "mistralai/Mistral-7B-Instruct-v0.1"
-
-# Load tokenizer
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+#hf_token = os.getenv("HF_TOKEN")
 
-# 🔧 Use 4-bit or 8-bit quantization
-bnb_config = BitsAndBytesConfig(
-    load_in_4bit=True,             # or use load_in_8bit=True
-    bnb_4bit_compute_dtype="float16",
-    bnb_4bit_use_double_quant=True,
-    bnb_4bit_quant_type="nf4",
-)
-
-# Load model with quantization
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    device_map="auto",
-    quantization_config=bnb_config,
-    token=hf_token
-)
 
-# Create pipeline
-pipe = pipeline(
-    "text-generation",
-    model=model,
-    tokenizer=tokenizer,
-    max_new_tokens=512,
+llm = HuggingFaceEndpoint(
+    repo_id="HuggingFaceH4/zephyr-7b-beta",
+    task="text-generation",
+    huggingfacehub_api_token=os.getenv("HF_TOKEN"),
     temperature=0.7,
-    top_p=0.95,
-    repetition_penalty=1.15
+    max_new_tokens=512
 )
 
-# LangChain wrapper
-llm = HuggingFacePipeline(pipeline=pipe)
-
 # Initialize LangChain agent
 agent = initialize_agent(
     tools=tools,