diff --git a/README.md b/README.md
index 7c7a6f243d6..4a2ea1a2426 100644
--- a/README.md
+++ b/README.md
@@ -177,10 +177,15 @@ Specify OpenAI organization ID.
 
 ### `AZURE_URL` (optional)
 
-> Example: https://{azure-resource-url}/openai/deployments/{deploy-name}
+> Example: https://{azure-resource-url}/openai/deployments/{deploy-name} 
+> 
+> Example: https://xxx.openai.azure.com/openai/deployments/{deploy-name}
 
 Azure deploy url.
 
+If `{deploy-name}` is using the template mode, then it will automatically replace the path based on the model selected by the client. 
+If your model name is different from the deployment name, then you need to set the `AZURE_OPENAI_MODEL_MAPPER` parameter.
+
 ### `AZURE_API_KEY` (optional)
 
 Azure Api Key.
@@ -216,12 +221,22 @@ If you want to disable parse settings from url, set this to 1.
 ### `CUSTOM_MODELS` (optional)
 
 > Default: Empty
+> 
 > Example: `+llama,+claude-2,-gpt-3.5-turbo,gpt-4-1106-preview=gpt-4-turbo` means add `llama, claude-2` to model list, and remove `gpt-3.5-turbo` from list, and display `gpt-4-1106-preview` as `gpt-4-turbo`.
+>
+> Example: `-all,gpt-35-turbo,gpt-4,gpt-4-32k`. The meaning is to only display `gpt-3.5-turbo`, `gpt-4`, and `gpt-4-32k` in the model list. 
 
 To control custom models, use `+` to add a custom model, use `-` to hide a model, use `name=displayName` to customize model name, separated by comma.
 
 User `-all` to disable all default models, `+all` to enable all default models.
 
+### `AZURE_OPENAI_MODEL_MAPPER` (optional)
+> Default: Empty
+> Example: `gpt-3.5-turbo=gpt-35-turbo` means map `gpt-3.5-turbo` to `gpt-35-turbo`
+
+If you are deploying ChatGPT using Azure OpenAI, it is recommended to set the `AZURE_OPENAI_MODEL_MAPPER`. 
+The session summarization feature relies on the `gpt-3.5-turbo` model, unless the name of your Azure deployment is the same as it.
+
 ## Requirements
 
 NodeJS >= 18, Docker >= 20
diff --git a/README_CN.md b/README_CN.md
index d734796581b..3cc6e4b95f7 100644
--- a/README_CN.md
+++ b/README_CN.md
@@ -124,9 +124,16 @@ Azure Api 版本，你可以在这里找到：[Azure 文档](https://learn.micro
 
 > 示例：`+qwen-7b-chat,+glm-6b,-gpt-3.5-turbo,gpt-4-1106-preview=gpt-4-turbo` 表示增加 `qwen-7b-chat` 和 `glm-6b` 到模型列表，而从列表中删除 `gpt-3.5-turbo`，并将 `gpt-4-1106-preview` 模型名字展示为 `gpt-4-turbo`。
 > 如果你想先禁用所有模型，再启用指定模型，可以使用 `-all,+gpt-3.5-turbo`，则表示仅启用 `gpt-3.5-turbo`
-
+> Example: `-all,gpt-35-turbo,gpt-4,gpt-4-32k`. 意思是在model 列表中只显示gpt-35-turbo,gpt-4,gpt-4-32k
 用来控制模型列表，使用 `+` 增加一个模型，使用 `-` 来隐藏一个模型，使用 `模型名=展示名` 来自定义模型的展示名，用英文逗号隔开。
 
+### `AZURE_OPENAI_MODEL_MAPPER` (optional)
+> Default: Empty
+> Example: `gpt-3.5-turbo=gpt-35-turbo` 意思是 `gpt-3.5-turbo` 映射到 `gpt-35-turbo`
+
+如果你使用azure openai 来部署的chatgpt,建议设定AZURE_OPENAI_MODEL_MAPPER . 
+会话摘要功能依赖gpt-3.5-turbo 这个model,除非你的azure deployment的名字和它相同
+
 ## 开发
 
 点击下方按钮，开始二次开发：
diff --git a/app/api/common.ts b/app/api/common.ts
index 6b0d619df1d..7edd96358f8 100644
--- a/app/api/common.ts
+++ b/app/api/common.ts
@@ -6,6 +6,13 @@ import { makeAzurePath } from "../azure";
 
 const serverConfig = getServerSideConfig();
 
+export function azureModelMaper(model: string) {
+  if (serverConfig.isAzure) {
+    return serverConfig.azureModelMaper[model] ?? model;
+  }
+  return model;
+}
+
 export async function requestOpenai(req: NextRequest) {
   const controller = new AbortController();
 
@@ -28,6 +35,13 @@ export async function requestOpenai(req: NextRequest) {
     baseUrl = baseUrl.slice(0, -1);
   }
 
+  let body = await req.json();
+  console.log("[model name]", body["model"]);
+  baseUrl = baseUrl.replace(
+    "{deploy-name}",
+    azureModelMaper(body["model"]) as string,
+  );
+
   console.log("[Proxy] ", path);
   console.log("[Base Url]", baseUrl);
   // this fix [Org ID] undefined in server side if not using custom point
@@ -63,7 +77,7 @@ export async function requestOpenai(req: NextRequest) {
       }),
     },
     method: req.method,
-    body: req.body,
+    body: JSON.stringify(body),
     // to fix #2485: https://stackoverflow.com/questions/55920957/cloudflare-worker-typeerror-one-time-use-body
     redirect: "manual",
     // @ts-ignore
@@ -72,7 +86,7 @@ export async function requestOpenai(req: NextRequest) {
   };
 
   // #1815 try to refuse gpt4 request
-  if (serverConfig.customModels && req.body) {
+  if (serverConfig.customModels && body) {
     try {
       const modelTable = collectModelTable(
         DEFAULT_MODELS,
diff --git a/app/config/server.ts b/app/config/server.ts
index 2398805a264..58779ff8122 100644
--- a/app/config/server.ts
+++ b/app/config/server.ts
@@ -70,6 +70,16 @@ export const getServerSideConfig = () => {
     `[Server Config] using ${randomIndex + 1} of ${apiKeys.length} api key`,
   );
 
+  let AzureModelMaper: Record<string, string> = {};
+  (process.env.AZURE_OPENAI_MODEL_MAPPER || "")
+    .trim()
+    .split(",")
+    .map((v) => v.trim().split("="))
+    .forEach(([k, v]) => {
+      AzureModelMaper[k] = v;
+    });
+  console.log("[AZURE_OPENAI_MODEL_MAPPER]", AzureModelMaper);
+
   return {
     baseUrl: process.env.BASE_URL,
     apiKey,
@@ -79,6 +89,7 @@ export const getServerSideConfig = () => {
     azureUrl: process.env.AZURE_URL,
     azureApiKey: process.env.AZURE_API_KEY,
     azureApiVersion: process.env.AZURE_API_VERSION,
+    azureModelMaper: AzureModelMaper,
 
     needCode: ACCESS_CODES.size > 0,
     code: process.env.CODE,
diff --git a/next.config.mjs b/next.config.mjs
index 4faa63e5450..c62e33307b7 100644
--- a/next.config.mjs
+++ b/next.config.mjs
@@ -24,6 +24,8 @@ const nextConfig = {
       child_process: false,
     };
 
+    // open source map
+    config.devtool = 'source-map';
     return config;
   },
   output: mode,