New model: Falcon7B

Falcon40B was too heavy

app.py

@@ -3,7 +3,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 import os
 
-MODEL_NAME = "tiiuae/falcon-40b"
+MODEL_NAME = "tiiuae/falcon-7b"
 
 access_token = os.getenv("HF_ACCESS_TOKEN")
 tokenizer = AutoTokenizer.from_pretrained(
@@ -43,7 +43,7 @@ def improve_code(code: str) -> str:
 app = gr.Blocks()
 
 with app:
-    gr.Markdown("## MCP Server Code Improver with Falcon 40B")
+    gr.Markdown("## MCP Server Code Improver with Falcon 7B")
     code_input = gr.Textbox(label="Original code", lines=15)
     improve_btn = gr.Button("Improve Code")
     code_output = gr.Textbox(label="Improved code", lines=15)