Update app.py

app.py

@@ -7,7 +7,7 @@ from threading import Thread
 # Remove GPU decorator since we are CPU-only
 def predict(message, history):
     # Load model and tokenizer on CPU
-    model_id = "kurakurai/Luth-LFM2-1.2B"
+    model_id = "kurakurai/Luth-LFM2-350M"
     tokenizer = AutoTokenizer.from_pretrained(model_id)
     model = AutoModelForCausalLM.from_pretrained(
         model_id,
@@ -59,9 +59,9 @@ def predict(message, history):
 gr.ChatInterface(
     predict,
     description="""
-    <center><h2>Kurakura AI Luth-LFM2-1.2B Chat</h2></center>
+    <center><h2>Kurakura AI Luth-LFM2-350M Chat</h2></center>
     
-    Chat with [Luth-LFM2-1.2B](https://huggingface.co/kurakurai/Luth-LFM2-1.2B), a French-tuned version of LFM2-1.2B.
+    Chat with [Luth-LFM2-350M](https://huggingface.co/kurakurai/Luth-LFM2-350M), a French-tuned version of LFM2-350M.
     """,
     examples=[
         "Peux-tu résoudre l'équation 3x - 7 = 11 pour x ?",