diff --git a/README.md b/README.md
index 9ef33c06a1ec1cf7a825cea26683f59aba08334e..7b3e6bb13afdd1692086fd2e6c28d97d1d06d237 100644
--- a/README.md
+++ b/README.md
@@ -101,6 +101,7 @@
 
    ```bash
    $ cd ragflow/docker
+   $ chmod +x ./entrypoint.sh
    $ docker compose up -d
    ```
 
@@ -165,12 +166,13 @@ $ git clone https://github.com/infiniflow/ragflow.git
 $ cd ragflow/
 $ docker build -t infiniflow/ragflow:v1.0 .
 $ cd ragflow/docker
+$ chmod +x ./entrypoint.sh
 $ docker compose up -d
 ```
 
 ## 🆕 Latest Features
 
-- Support [Ollam](./docs/ollama.md) for local LLM deployment.
+- Support [Ollama](./docs/ollama.md) for local LLM deployment.
 - Support Chinese UI.
 
 ## đź“ś Roadmap
diff --git a/README_ja.md b/README_ja.md
index 8437bebfb7719985b5d430f501b290e525a481f1..2c5f50f6018ed9dc3379c320eab802298b2a49a6 100644
--- a/README_ja.md
+++ b/README_ja.md
@@ -101,6 +101,7 @@
 
    ```bash
    $ cd ragflow/docker
+   $ chmod +x ./entrypoint.sh
    $ docker compose up -d
    ```
 
@@ -165,12 +166,13 @@ $ git clone https://github.com/infiniflow/ragflow.git
 $ cd ragflow/
 $ docker build -t infiniflow/ragflow:v1.0 .
 $ cd ragflow/docker
+$ chmod +x ./entrypoint.sh
 $ docker compose up -d
 ```
 
 ## 🆕 最新の新機能
 
-- [Ollam](./docs/ollama.md) を使用した大規模モデルのローカライズされたデプロイメントをサポートします。
+- [Ollama](./docs/ollama.md) を使用した大規模モデルのローカライズされたデプロイメントをサポートします。
 - 中国語インターフェースをサポートします。
 
 ## 📜 ロードマップ
diff --git a/README_zh.md b/README_zh.md
index eec642e8adeced8c5de6814eff192e3a65e981cd..21c93cdda10bd47c453973e41869a8f1ad7e61d7 100644
--- a/README_zh.md
+++ b/README_zh.md
@@ -101,6 +101,7 @@
 
    ```bash
    $ cd ragflow/docker
+   $ chmod +x ./entrypoint.sh
    $ docker compose -f docker-compose-CN.yml up -d
    ```
 
@@ -165,12 +166,13 @@ $ git clone https://github.com/infiniflow/ragflow.git
 $ cd ragflow/
 $ docker build -t infiniflow/ragflow:v1.0 .
 $ cd ragflow/docker
+$ chmod +x ./entrypoint.sh
 $ docker compose up -d
 ```
 
 ## 🆕 最近新特性
 
-- 支持用 [Ollam](./docs/ollama.md) 对大模型进行本地化部署。
+- 支持用 [Ollama](./docs/ollama.md) 对大模型进行本地化部署。
 - 支持中文界面。
 
 ## 📜 路线图
diff --git a/api/apps/conversation_app.py b/api/apps/conversation_app.py
index 42339e1f00dae05b0202b15b976b17e8ef42885e..6ece253cafa969aedba934da74692b2148837cfe 100644
--- a/api/apps/conversation_app.py
+++ b/api/apps/conversation_app.py
@@ -20,7 +20,7 @@ from flask_login import login_required
 from api.db.services.dialog_service import DialogService, ConversationService
 from api.db import LLMType
 from api.db.services.knowledgebase_service import KnowledgebaseService
-from api.db.services.llm_service import LLMService, LLMBundle
+from api.db.services.llm_service import LLMService, LLMBundle, TenantLLMService
 from api.settings import access_logger, stat_logger, retrievaler, chat_logger
 from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
 from api.utils import get_uuid
@@ -184,8 +184,11 @@ def chat(dialog, messages, **kwargs):
     assert messages[-1]["role"] == "user", "The last content of this conversation is not from user."
     llm = LLMService.query(llm_name=dialog.llm_id)
     if not llm:
-        raise LookupError("LLM(%s) not found" % dialog.llm_id)
-    llm = llm[0]
+        llm = TenantLLMService.query(tenant_id=dialog.tenant_id, llm_name=dialog.llm_id)
+        if not llm:
+            raise LookupError("LLM(%s) not found" % dialog.llm_id)
+        max_tokens = 1024
+    else: max_tokens = llm[0].max_tokens
     questions = [m["content"] for m in messages if m["role"] == "user"]
     embd_mdl = LLMBundle(dialog.tenant_id, LLMType.EMBEDDING)
     chat_mdl = LLMBundle(dialog.tenant_id, LLMType.CHAT, dialog.llm_id)
@@ -227,11 +230,11 @@ def chat(dialog, messages, **kwargs):
     gen_conf = dialog.llm_setting
     msg = [{"role": m["role"], "content": m["content"]}
            for m in messages if m["role"] != "system"]
-    used_token_count, msg = message_fit_in(msg, int(llm.max_tokens * 0.97))
+    used_token_count, msg = message_fit_in(msg, int(max_tokens * 0.97))
     if "max_tokens" in gen_conf:
         gen_conf["max_tokens"] = min(
             gen_conf["max_tokens"],
-            llm.max_tokens - used_token_count)
+            max_tokens - used_token_count)
     answer = chat_mdl.chat(
         prompt_config["system"].format(
             **kwargs), msg, gen_conf)