Update run.sh

run.sh

@@ -9,11 +9,11 @@ sleep 5
 # Pull the model from Hugging Face.
 # This command downloads the model and makes it available to the API.
 # It's the non-interactive version of "ollama run".
+ollama pull hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M
 ollama pull hf.co/unsloth/gemma-3-4b-it-qat-GGUF:Q4_K_M
-ollama pull hf.co/Menlo/Jan-nano-128k-gguf:Q4_K_M
-ollama pull hf.co/bartowski/Qwen_Qwen3-4B-GGUF:Q4_K_M
 ollama pull hf.co/bartowski/Qwen_Qwen3-1.7B-GGUF:Q5_K_M
-ollama pull hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M
+ollama pull gemma3n:e2b-it-q4_K_M
+ollama pull granite3.3:2b
 
 # Start the Gradio web application
 # This will connect to the Ollama server which is already running.