change models

app.py

@@ -10,29 +10,23 @@ from llama_cpp_agent.chat_history.messages import Roles
 import gradio as gr
 from huggingface_hub import hf_hub_download
 
-
 huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
 
 hf_hub_download(
-    repo_id="bartowski/gemma-2-9b-it-GGUF",
-    filename="gemma-2-9b-it-Q5_K_M.gguf",
-    local_dir="./models"
-)
-
-hf_hub_download(
-    repo_id="bartowski/gemma-2-27b-it-GGUF",
-    filename="gemma-2-27b-it-Q5_K_M.gguf",
+    repo_id="Qwen/Qwen2-0.5B-Instruct-GGUF",
+    filename="qwen2-0_5b-instruct-q4_k_m.gguf",
     local_dir="./models"
 )
 
-hf_hub_download(
-    repo_id="google/gemma-2-2b-it-GGUF",
-    filename="2b_it_v2.gguf",
-    local_dir="./models",
-    token=huggingface_token
-)
-
+#hf_hub_download(
+#    repo_id="TobDeBer/Meta-Llama-3.1-8B-Instruct-Q4_K_M-GGUF",
+#    filename="meta-llama-3.1-8b-instruct-q4_k_m.gguf",
+#    local_dir="./models",
+#    token=huggingface_token
+#)
 
+# TobDeBer/granite-8b-code-instruct-128k-Q4_K_M-GGUF
+# granite-8b-code-instruct-128k-q4_k_m.gguf
 
 llm = None
 llm_model = None
@@ -108,26 +102,17 @@ def respond(
         outputs += output
         yield outputs
 
-description = """<p align="center">Defaults to 2B (you can switch to 9B or 27B from additional inputs)</p>
-<p><center>
-<a href="https://huggingface.co/google/gemma-2-27b-it" target="_blank">[27B it Model]</a>
-<a href="https://huggingface.co/google/gemma-2-9b-it" target="_blank">[9B it Model]</a>
-<a href="https://huggingface.co/google/gemma-2-2b-it" target="_blank">[2B it Model]</a>
-<a href="https://huggingface.co/bartowski/gemma-2-27b-it-GGUF" target="_blank">[27B it Model GGUF]</a>
-<a href="https://huggingface.co/bartowski/gemma-2-9b-it-GGUF" target="_blank">[9B it Model GGUF]</a>
-<a href="https://huggingface.co/google/gemma-2-2b-it-GGUF" target="_blank">[2B it Model GGUF]</a>
-</center></p>
+description = """<p align="center">Defaults Qwen 500M</p>
 """
 
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
         gr.Dropdown([
-                'gemma-2-9b-it-Q5_K_M.gguf',
-                'gemma-2-27b-it-Q5_K_M.gguf',
-                '2b_it_v2.gguf'
+                'meta-llama-3.1-8b-instruct-q4_k_m.gguf',
+                'qwen2-0_5b-instruct-q4_k_m.gguf'
             ],
-            value="2b_it_v2.gguf",
+            value="qwen2-0_5b-instruct-q4_k_m.gguf",
             label="Model"
         ),
         gr.Textbox(value="You are a helpful assistant.", label="System message"),
@@ -159,7 +144,7 @@ demo = gr.ChatInterface(
     undo_btn="Undo",
     clear_btn="Clear",
     submit_btn="Send",
-    title="Chat with Gemma 2 using llama.cpp", 
+    title="Chat with Qwen 2 using llama.cpp", 
     description=description,
     chatbot=gr.Chatbot(
         scale=1, 

requirements.txt

@@ -1,4 +1,5 @@
-huggingface_hub==0.22.2
+spaces
+huggingface_hub
 scikit-build-core
 https://github.com/abetlen/llama-cpp-python/releases/download/v0.2.81-cu124/llama_cpp_python-0.2.81-cp310-cp310-linux_x86_64.whl
-llama-cpp-agent>=0.2.25
+llama-cpp-agent>=0.2.25