switch to 1.5b

app.py

@@ -3,8 +3,8 @@ from llama_cpp import Llama
 from huggingface_hub import hf_hub_download
 
 hf_hub_download(
-    repo_id="Qwen/Qwen2.5-3B-Instruct-GGUF",
-    filename="qwen2.5-3b-instruct-q4_k_m.gguf",
+    repo_id="Qwen/Qwen2.5-1.5B-Instruct-GGUF",
+    filename="qwen2.5-1.5b-instruct-q4_k_m.gguf",
     local_dir="./models",
 )
 
@@ -12,7 +12,7 @@ hf_hub_download(
 @st.cache_resource
 def load_model():
     return Llama(
-        model_path="models/qwen2.5-3b-instruct-q4_k_m.gguf",
+        model_path="models/qwen2.5-1.5b-instruct-q4_k_m.gguf",
         n_ctx=2048,
         n_threads=6,
         n_batch=8,
@@ -28,7 +28,7 @@ llm = load_model()
 if "chat_history" not in st.session_state:
     st.session_state.chat_history = []
 
-st.title("🧠 Qwen2.5-3B-Instruct (Streamlit + GGUF)")
+st.title("🧠 Qwen2.5-1.5B-Instruct (Streamlit + GGUF)")
 st.caption("Powered by `llama.cpp` and `llama-cpp-python` | 4-bit Q4_K_M inference")
 
 with st.sidebar: