diff --git a/README.md b/README.md
index 94e6054..11f5858 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,3 @@
 # ComfyUI-YCYY-API
-Call API in ComfyUI
+Freely call APIs in ComfyUI
+
diff --git a/gemini_image_node.py b/gemini_image_node.py
index 17f7e22..6adf11b 100644
--- a/gemini_image_node.py
+++ b/gemini_image_node.py
@@ -206,7 +206,7 @@ class GeminiImage(io.ComfyNode):
             return cls._parse_response(resp)
         except Exception as e:
             empty_image = cls._create_empty_image()
-            return io.NodeOutput(empty_image,"API请求失败，请检查接口地址和 KEY 是否正确")    
+            return io.NodeOutput(empty_image,f'{{"success":false,"message":"The API request failed. Please check if the interface address and key are correct."}}')    
     # 将图像张量输入转换为与 Gemini API 兼容的格式。
     @classmethod
     def _create_image_parts(cls,image_input):
@@ -257,24 +257,24 @@ class GeminiImage(io.ComfyNode):
             return cls._parse_response(resp)
         except Exception as e:
             empty_image = cls._create_empty_image()
-            return io.NodeOutput(empty_image,"API请求失败，请检查接口地址和 KEY 是否正确")
+            return io.NodeOutput(empty_image,f'{{"success":false,"message":"The API request failed. Please check if the interface address and key are correct."}}')
     # 解析response 返回内容
     @classmethod
     def _parse_response(cls,resp):
         # 检查HTTP状态码
         if resp.status_code != 200:
             empty_image = cls._create_empty_image()
-            return (empty_image,f"API请求返回错误 (状态码: {resp.status_code}):(错误原因:{resp.text})")
+            return (empty_image,f'{{"success":false,"message":"API request returns an error.status_code:{resp.status_code}.error_reason:{resp.text}"}}')
         # 检查返回内容是否为空
         if not resp.text.strip():
             empty_image = cls._create_empty_image()
-            return (empty_image,"API返回内容为空")
+            return (empty_image,f'{{"success":false,"message":"The API returns an empty content"}}')
         try:
             data = resp.json()
         except Exception as json_exception:
             # print(f"JSON解析失败：{json_exception}")
             empty_image = cls._create_empty_image()
-            return (empty_image,"API返回JSON解析失败")
+            return (empty_image,f'{{"success":false,"message":"The API returned a JSON parsing failure"}}')
         # 解析响应数据
         if "candidates" in data and data["candidates"]:
             candidate = data["candidates"][0]
@@ -295,7 +295,7 @@ class GeminiImage(io.ComfyNode):
         else:
             # print(f"未找到imag数据")
             empty_image = cls._create_empty_image()
-            return (empty_image,"未找到imag数据")
+            return (empty_image,f'{{"success":false,"message":"Imag data not found"}}')
     # 将返回的图像数据解析为 comfyui 格式的 image
     @classmethod
     def _image_data_to_comfyui_image(cls,image_date):
@@ -307,14 +307,14 @@ class GeminiImage(io.ComfyNode):
             return image_tensor
         except Exception as e:
             empty_image = cls._create_empty_image()
-            return io.NodeOutput(empty_image,"返回图像数据解析异常")
+            return io.NodeOutput(empty_image,f'{{"success":false,"message":"Return image data parsing exception"}}')
     # 获取token用量
     @classmethod
     def _format_tokens_usage(cls,usageMetadata):
         if not usageMetadata:
             return ""
         total_tokens = usageMetadata.get('totalTokenCount', '-')
-        return f"{{\"success\":true,\"total_tokens\":{total_tokens}}}"
+        return f'{{"success":true,"message":"total_tokens:{total_tokens}"}}'
     # 创建空图像
     @classmethod
     def _create_empty_image(cls):
diff --git a/locales/en/nodeDefs.json b/locales/en/nodeDefs.json
new file mode 100644
index 0000000..47043dd
--- /dev/null
+++ b/locales/en/nodeDefs.json
@@ -0,0 +1,33 @@
+{
+    "YCYY_Gemini_Image_API": {
+        "display_name": "YCYY Gemini Image API",
+        "inputs": {
+            "images": {
+                "name": "image",
+                "tooltip": "Optional image(s) to use as context for the model"
+            },
+            "prompt": {
+                "name": "prompt"
+            },
+            "model": {
+                "name": "model"
+            },
+            "aspectRatio": {
+                "name": "aspectRatio",
+                "tooltip": "The model defaults to matching the output image size to that of your input image, or otherwise generates 1:1 squares. You can control the aspect ratio of the output image using the aspect ratio"
+            },
+            "seed": {
+                "name": "seed",
+                "tooltip": "Seed to use for generation"
+            }
+        },
+        "outputs": {
+            "0": {
+                "name": "Image"
+            },
+            "1": {
+                "name": "String"
+            }
+        }
+    }
+}
\ No newline at end of file
diff --git a/locales/zh/nodeDefs.json b/locales/zh/nodeDefs.json
new file mode 100644
index 0000000..0500ce2
--- /dev/null
+++ b/locales/zh/nodeDefs.json
@@ -0,0 +1,33 @@
+{
+    "YCYY_Gemini_Image_API": {
+        "display_name": "浥尘 Gemini 图像 API",
+        "inputs": {
+            "images": {
+                "name": "图像",
+                "tooltip": "可选的图像，用作模型的上下文"
+            },
+            "prompt": {
+                "name": "prompt"
+            },
+            "model": {
+                "name": "模型"
+            },
+            "aspectRatio": {
+                "name": "图像宽高比",
+                "tooltip": "该模型默认将输出图像大小与输入图像大小匹配，否则生成 1:1 正方形。您可以使用宽高比来控制输出图像的宽高比。"
+            },
+            "seed": {
+                "name": "种子",
+                "tooltip": "用于生成噪声的随机种子。"
+            }
+        },
+        "outputs": {
+            "0": {
+                "name": "图像"
+            },
+            "1": {
+                "name": "字符串"
+            }
+        }
+    }
+}
\ No newline at end of file