Update app.py

app.py

@@ -1,5 +1,5 @@
 import gradio as gr
-from transformers import pipeline, BitsAndBytesConfig
+from transformers import pipeline, BitsAndBytesConfig, AutoModelForCausalLM, AutoTokenizer
 import PyPDF2
 import io
 import re
@@ -21,15 +21,24 @@ quant_config = BitsAndBytesConfig(
     bnb_4bit_quant_type="nf4"
 )
 
-# Initialize pipeline
-analyzer = pipeline(
-    "text-generation",
-    model="mistralai/Mistral-7B-Instruct-v0.3",
+# Load tokenizer and model with quantization
+tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.3")
+model = AutoModelForCausalLM.from_pretrained(
+    "mistralai/Mistral-7B-Instruct-v0.3",
     device_map="auto",
     quantization_config=quant_config,
     torch_dtype=torch.float16
 )
 
+# Initialize pipeline with preloaded model and tokenizer
+analyzer = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    device_map="auto",  # Still respected from model
+    torch_dtype=torch.float16
+)
+
 # Skills set for faster lookups
 GENERAL_SKILLS = {
     'communication', 'problem solving', 'project management',