max new tokens

ai_med_extract/api/routes.py

@@ -388,7 +388,7 @@ def register_routes(app, agents):
             import torch
             torch.set_num_threads(2)
             inputs = pipeline.tokenizer([prompt], return_tensors="pt")
-            outputs = pipeline.model.generate(**inputs, max_new_tokens=1000, do_sample=False, pad_token_id=pipeline.tokenizer.eos_token_id or 32000)
+            outputs = pipeline.model.generate(**inputs, max_new_tokens=4000, do_sample=False, pad_token_id=pipeline.tokenizer.eos_token_id or 32000)
             text = pipeline.tokenizer.decode(outputs[0], skip_special_tokens=True)
             
             # Robust summary extraction with fallback
@@ -1250,7 +1250,7 @@ def register_routes(app, agents):
             if not pipeline:
                 return jsonify({"error": "Model pipeline not available"}), 500
             inputs = pipeline.tokenizer([prompt], return_tensors="pt")
-            outputs = pipeline.model.generate(**inputs, max_new_tokens=1000, do_sample=False, pad_token_id=pipeline.tokenizer.eos_token_id or 32000)
+            outputs = pipeline.model.generate(**inputs, max_new_tokens=4000, do_sample=False, pad_token_id=pipeline.tokenizer.eos_token_id or 32000)
             text = pipeline.tokenizer.decode(outputs[0], skip_special_tokens=True)
             new_summary = text.split("Now generate the complete, updated clinical summary with all four sections in a markdown format:")[-1].strip()
             # For other models, after extracting new_summary: