add app.py

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+from llama_cpp import Llama
+import os 
+model_file = "Yi-6B.q4_k_m.gguf"
+if not os.path.isfile("Yi-6B.q4_k_m.gguf"):
+    os.system("wget -c https://huggingface.co/SamPurkis/Yi-6B-GGUF/resolve/main/Yi-6B.q4_k_m.gguf")
+
+llm = Llama(model_path=model_file)
+
+def generate_text(input_text):
+    output = llm(f"Human: {input_text} A:", max_tokens=512, stop=["Assistant:", "\n"], echo=True)
+    return output['choices'][0]['text']
+
+input_text = gr.inputs.Textbox(lines= 10, label="Enter your input text")
+output_text = gr.outputs.Textbox(label="Output text")
+
+description = "llama.cpp implementation in python [https://github.com/abetlen/llama-cpp-python]"
+
+examples = [
+    ["What is the capital of France? ", "The capital of France is Paris."],
+    ["Who wrote the novel 'Pride and Prejudice'?", "The novel 'Pride and Prejudice' was written by Jane Austen."],
+    ["What is the square root of 64?", "The square root of 64 is 8."]
+]
+
+gr.Interface(fn=generate_text, inputs=input_text, outputs=output_text, title="Llama Language Model", description=description, examples=examples).launch()

requirements.txt

@@ -0,0 +1,2 @@
+llama_cpp
+gradio