Upload app.py

app.py

@@ -3,15 +3,15 @@ from llama_cpp import Llama
 
 llm = Llama(model_path="vicuna-AlekseyKorshuk-7B-GPTQ-4bit-128g.GGML.bin", n_ctx=2048)
 
-def generate_text(llm, prompt, max_tokens=468, temperature=0.1, top_p=0.5):
-    output = llm(prompt, max_tokens=max_tokens, temperature=temperature, top_p=top_p, echo=False, stop=["#"])
+def generate_text(prompt):
+    output = llm(prompt, max_tokens=468, temperature=0.1, top_p=0.5, echo=False, stop=["#"])
     text = output['choices'][0]['text']
     return 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]"
+description = "Vicuna-7B-GPTQ-4bit-128g.GGML, max_tokens=468, temperature=0.1, top_p=0.5"
 
 examples = [
     ["What is the capital of France? ", "The capital of France is Paris."],