Update app.py

app.py

@@ -45,8 +45,8 @@ def build_prompt(req1, req2, prompt_type="zero-shot"):
 # -----------------------------
 @app.on_event("startup")
 def load_models():
-    print("Loading DeepSeek model into memory...")
-    deepseek_name = "deepseek-ai/DeepSeek-R1-Distill-Llama-8B"
+    print("Loading smaller DeepSeek model into memory...")
+    deepseek_name = "deepseek-ai/deepseek-vl2-small"   # smaller model
     app.state.deepseek_tokenizer = AutoTokenizer.from_pretrained(deepseek_name)
     app.state.deepseek_tokenizer.pad_token = app.state.deepseek_tokenizer.eos_token
     app.state.deepseek_model = AutoModelForCausalLM.from_pretrained(
@@ -68,16 +68,12 @@ def run_gpt4(req1, req2, prompt_type, api_key):
     )
     return completion.choices[0].message.content.strip()
 
+
 def run_deepseek(req1, req2, prompt_type):
     tokenizer = app.state.deepseek_tokenizer
     model = app.state.deepseek_model
     prompt = build_prompt(req1, req2, prompt_type)
-    inputs = tokenizer(
-        [prompt],
-        return_tensors="pt",
-        padding=True,
-        truncation=True
-    )
+    inputs = tokenizer([prompt], return_tensors="pt", padding=True, truncation=True)
     outputs = model.generate(
         input_ids=inputs.input_ids,
         attention_mask=inputs.attention_mask,