ui

app.py

@@ -15,6 +15,20 @@ model = AutoModelForCausalLM.from_pretrained(
     model_id, gguf_file=filename, device_map="auto"
 )
 
+example = """int __fastcall sub_B0D04(int a1, int a2)
+{
+  unsigned int v2; // r4
+  int result; // r0
+
+  v2 = a1 + a2;
+  if ( __CFADD__(a1, a2) )
+    return 0;
+  result = _libc_alloca_cutoff();
+  if ( v2 <= 0x1000 )
+    return result | 1;
+  return result;
+}"""
+
 # Then create the pipeline with the model and tokenizer
 pipe = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
 
@@ -57,8 +71,8 @@ def predict(code):
 
 demo = gr.Interface(
     fn=predict,
-    inputs="text",
-    outputs="text",
+    inputs=gr.Text(placeholder=example, label="Hex-Rays decompiler output"),
+    outputs=gr.JSON(label="Aidapal Output"),,
     description=frontmatter.load("README.md").content,
 )
 demo.launch()