inference model

app.py

@@ -1,10 +1,10 @@
 import gradio as gr
 from llama_cpp import Llama
 
-llm = Llama(model_path="ggml-alpaca-7b-q4.bin", n_ctx=256, n_batch=128)
+llm = Llama(model_path="ggml-model-q4_0.bin")
 
 def generate_text(input_text):
-    output = llm(f"{input_text}", max_tokens=128, stop=["Q:", "\n", "#"], echo=False)
+    output = llm(f"### Instruction:\n{input_text}\n\n### Response:\n", max_tokens=256, echo=False)
     return output['choices'][0]['text']
 
 input_text = gr.inputs.Textbox(lines= 10, label="Enter your input text")

ggml-model-q4_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84905e755a89828b935fc0f1031a28bc2199c7710095fe3dbe64e8afee5f187b
+size 3791725184