Deploy Auto-Quantization MVP

quantizer.py

@@ -74,7 +74,7 @@ async def quantize_model(job: Dict) -> Dict:
         # Step 2: Load tokenizer
         print(f"\n📋 Step 2/5: Loading tokenizer...")
         try:
-            tokenizer = AutoTokenizer.from_pretrained(model_id)
+            tokenizer = AutoTokenizer.from_pretrained(model_id, token=HF_TOKEN)
             print(f"✓ Tokenizer loaded")
         except Exception as e:
             raise Exception(f"Failed to load tokenizer: {str(e)}")
@@ -98,7 +98,8 @@ async def quantize_model(job: Dict) -> Dict:
                 quantization_config=quant_config,
                 torch_dtype=torch.float16,
                 low_cpu_mem_usage=True,
-                trust_remote_code=False  # Security: don't trust remote code
+                trust_remote_code=False,  # Security: don't trust remote code
+                token=HF_TOKEN
             )
             print(f"✓ Model quantized successfully")