diff --git a/README.md b/README.md
index 3f759dc..0b9577d 100644
--- a/README.md
+++ b/README.md
@@ -21,119 +21,8 @@
 
 llm模型导出`onnx`和`mnn`模型请使用[llm-export](https://github.com/wangzhaode/llm-export)
 
-`modelscope`模型下载：
+[模型下载](./docs/download.md)
 
-<details>
-  <summary>qwen</summary>
-
-- [modelscope-qwen-1.8b-chat]
-- [modelscope-qwen-7b-chat]
-- [modelscope-qwen-vl-chat]
-- [modelscope-qwen1.5-0.5b-chat]
-- [modelscope-qwen1.5-1.8b-chat]
-- [modelscope-qwen1.5-4b-chat]
-- [modelscope-qwen1.5-7b-chat]
-- [modelscope-qwen2-0.5b-instruct]
-- [modelscope-qwen2-1.5b-instruct]
-- [modelscope-qwen2-7b-instruct]
-- [modelscope-qwen2-vl-2b-instruct]
-- [modelscope-qwen2-vl-7b-instruct]
-- [modelscope-qwen2.5-0.5b-instruct]
-- [modelscope-qwen2.5-1.5b-instruct]
-- [modelscope-qwen2.5-3b-instruct]
-- [modelscope-qwen2.5-7b-instruct]
-- [modelscope-qwen2.5-coder-1.5b-instruct]
-- [modelscope-qwen2.5-coder-7b-instruct]
-- [modelscope-qwen2.5-math-1.5b-instruct]
-- [modelscope-qwen2.5-math-7b-instruct]
-- [modelscope-reader-lm-0.5b]
-- [modelscope-reader-lm-1.5b]
-
-</details>
-
-<details>
-  <summary>glm</summary>
-
-- [modelscope-chatglm-6b]
-- [modelscope-chatglm2-6b]
-- [modelscope-codegeex2-6b]
-- [modelscope-chatglm3-6b]
-- [modelscope-glm4-9b-chat]
-
-</details>
-
-<details>
-  <summary>llama</summary>
-
-- [modelscope-llama2-7b-chat]
-- [modelscope-llama3-8b-instruct]
-- [modelscope-llama3.2-1b-instruct]
-- [modelscope-llama3.2-3b-instruct]
-- [modelscope-baichuan2-7b-chat]
-- [modelscope-internlm-chat-7b]
-- [modelscope-yi-6b-chat]
-- [modelscope-deepseek-7b-chat]
-- [modelscope-tinyllama-1.1b-chat]
-
-</details>
-
-<details>
-  <summary>phi</summary>
-
-- [modelscope-phi-2]
-
-</details>
-
-<details>
-  <summary>embedding</summary>
-
-- [modelscope-bge-large-zh]
-- [modelscope-gte_sentence-embedding_multilingual-base]
-
-</details>
-
-[modelscope-qwen-1.8b-chat]: https://modelscope.cn/models/zhaode/Qwen-1_8B-Chat-MNN/files
-[modelscope-qwen-7b-chat]: https://modelscope.cn/models/zhaode/Qwen-7B-Chat-MNN/files
-[modelscope-qwen-vl-chat]: https://modelscope.cn/models/zhaode/Qwen-VL-Chat-MNN/files
-[modelscope-qwen1.5-0.5b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-0.5B-Chat-MNN/files
-[modelscope-qwen1.5-1.8b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-1.8B-Chat-MNN/files
-[modelscope-qwen1.5-4b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-4B-Chat-MNN/files
-[modelscope-qwen1.5-7b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-7B-Chat-MNN/files
-[modelscope-qwen2-0.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-0.5B-Instruct-MNN/files
-[modelscope-qwen2-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-1.5B-Instruct-MNN/files
-[modelscope-qwen2-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-7B-Instruct-MNN/files
-[modelscope-qwen2-vl-2b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-VL-2B-Instruct-MNN/files
-[modelscope-qwen2-vl-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-VL-7B-Instruct-MNN/files
-[modelscope-qwen2.5-0.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-0.5B-Instruct-MNN/files
-[modelscope-qwen2.5-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-3b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-3B-Instruct-MNN/files
-[modelscope-qwen2.5-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-7B-Instruct-MNN/files
-[modelscope-qwen2.5-coder-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Coder-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-coder-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Coder-7B-Instruct-MNN/files
-[modelscope-qwen2.5-math-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Math-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-math-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Math-7B-Instruct-MNN/files
-[modelscope-reader-lm-0.5b]: https://modelscope.cn/models/zhaode/reader-lm-0.5b-MNN/files
-[modelscope-reader-lm-1.5b]: https://modelscope.cn/models/zhaode/reader-lm-1.5b-MNN/files
-
-[modelscope-chatglm-6b]: https://modelscope.cn/models/zhaode/chatglm-6b-MNN/files
-[modelscope-chatglm2-6b]: https://modelscope.cn/models/zhaode/chatglm2-6b-MNN/files
-[modelscope-codegeex2-6b]: https://modelscope.cn/models/zhaode/codegeex2-6b-MNN/files
-[modelscope-chatglm3-6b]: https://modelscope.cn/models/zhaode/chatglm3-6b-MNN/files
-[modelscope-glm4-9b-chat]: https://modelscope.cn/models/zhaode/glm-4-9b-chat-MNN/files
-
-[modelscope-llama2-7b-chat]: https://modelscope.cn/models/zhaode/Llama-2-7b-chat-MNN/files
-[modelscope-llama3-8b-instruct]: https://modelscope.cn/models/zhaode/Llama-3-8B-Instruct-MNN/files
-[modelscope-llama3.2-1b-instruct]: https://modelscope.cn/models/zhaode/Llama-3.2-1B-Instruct-MNN/files
-[modelscope-llama3.2-3b-instruct]: https://modelscope.cn/models/zhaode/Llama-3.2-3B-Instruct-MNN/files
-[modelscope-baichuan2-7b-chat]: https://modelscope.cn/models/zhaode/Baichuan2-7B-Chat-MNN/files
-[modelscope-internlm-chat-7b]: https://modelscope.cn/models/zhaode/internlm-chat-7b-MNN/files
-[modelscope-yi-6b-chat]: https://modelscope.cn/models/zhaode/Yi-6B-Chat-MNN/files
-[modelscope-deepseek-7b-chat]: https://modelscope.cn/models/zhaode/deepseek-llm-7b-chat-MNN/files
-[modelscope-tinyllama-1.1b-chat]: https://modelscope.cn/models/zhaode/TinyLlama-1.1B-Chat-MNN/files
-[modelscope-phi-2]: https://modelscope.cn/models/zhaode/phi-2-MNN/files
-
-[modelscope-bge-large-zh]: https://modelscope.cn/models/zhaode/bge-large-zh-MNN/files
-[modelscope-gte_sentence-embedding_multilingual-base]: https://modelscope.cn/models/zhaode/gte_sentence-embedding_multilingual-base-MNN/files
 
 ## 构建
 
diff --git a/README_en.md b/README_en.md
index d4f6c17..be7abb6 100644
--- a/README_en.md
+++ b/README_en.md
@@ -21,119 +21,7 @@
 
 For exporting the llm model to `ONNX` or `mnn`, please use[llm-export](https://github.com/wangzhaode/llm-export)
 
-Download models from `modelscope`：
-
-<details>
-  <summary>qwen</summary>
-
-- [modelscope-qwen-1.8b-chat]
-- [modelscope-qwen-7b-chat]
-- [modelscope-qwen-vl-chat]
-- [modelscope-qwen1.5-0.5b-chat]
-- [modelscope-qwen1.5-1.8b-chat]
-- [modelscope-qwen1.5-4b-chat]
-- [modelscope-qwen1.5-7b-chat]
-- [modelscope-qwen2-0.5b-instruct]
-- [modelscope-qwen2-1.5b-instruct]
-- [modelscope-qwen2-7b-instruct]
-- [modelscope-qwen2-vl-2b-instruct]
-- [modelscope-qwen2-vl-7b-instruct]
-- [modelscope-qwen2.5-0.5b-instruct]
-- [modelscope-qwen2.5-1.5b-instruct]
-- [modelscope-qwen2.5-3b-instruct]
-- [modelscope-qwen2.5-7b-instruct]
-- [modelscope-qwen2.5-coder-1.5b-instruct]
-- [modelscope-qwen2.5-coder-7b-instruct]
-- [modelscope-qwen2.5-math-1.5b-instruct]
-- [modelscope-qwen2.5-math-7b-instruct]
-- [modelscope-reader-lm-0.5b]
-- [modelscope-reader-lm-1.5b]
-
-</details>
-
-<details>
-  <summary>glm</summary>
-
-- [modelscope-chatglm-6b]
-- [modelscope-chatglm2-6b]
-- [modelscope-codegeex2-6b]
-- [modelscope-chatglm3-6b]
-- [modelscope-glm4-9b-chat]
-
-</details>
-
-<details>
-  <summary>llama</summary>
-
-- [modelscope-llama2-7b-chat]
-- [modelscope-llama3-8b-instruct]
-- [modelscope-llama3.2-1b-instruct]
-- [modelscope-llama3.2-3b-instruct]
-- [modelscope-baichuan2-7b-chat]
-- [modelscope-internlm-chat-7b]
-- [modelscope-yi-6b-chat]
-- [modelscope-deepseek-7b-chat]
-- [modelscope-tinyllama-1.1b-chat]
-
-</details>
-
-<details>
-  <summary>phi</summary>
-
-- [modelscope-phi-2]
-
-</details>
-
-<details>
-  <summary>embedding</summary>
-
-- [modelscope-bge-large-zh]
-- [modelscope-gte_sentence-embedding_multilingual-base]
-
-</details>
-
-[modelscope-qwen-1.8b-chat]: https://modelscope.cn/models/zhaode/Qwen-1_8B-Chat-MNN/files
-[modelscope-qwen-7b-chat]: https://modelscope.cn/models/zhaode/Qwen-7B-Chat-MNN/files
-[modelscope-qwen-vl-chat]: https://modelscope.cn/models/zhaode/Qwen-VL-Chat-MNN/files
-[modelscope-qwen1.5-0.5b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-0.5B-Chat-MNN/files
-[modelscope-qwen1.5-1.8b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-1.8B-Chat-MNN/files
-[modelscope-qwen1.5-4b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-4B-Chat-MNN/files
-[modelscope-qwen1.5-7b-chat]: https://modelscope.cn/models/zhaode/Qwen1.5-7B-Chat-MNN/files
-[modelscope-qwen2-0.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-0.5B-Instruct-MNN/files
-[modelscope-qwen2-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-1.5B-Instruct-MNN/files
-[modelscope-qwen2-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-7B-Instruct-MNN/files
-[modelscope-qwen2-vl-2b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-VL-2B-Instruct-MNN/files
-[modelscope-qwen2-vl-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2-VL-7B-Instruct-MNN/files
-[modelscope-qwen2.5-0.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-0.5B-Instruct-MNN/files
-[modelscope-qwen2.5-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-3b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-3B-Instruct-MNN/files
-[modelscope-qwen2.5-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-7B-Instruct-MNN/files
-[modelscope-qwen2.5-coder-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Coder-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-coder-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Coder-7B-Instruct-MNN/files
-[modelscope-qwen2.5-math-1.5b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Math-1.5B-Instruct-MNN/files
-[modelscope-qwen2.5-math-7b-instruct]: https://modelscope.cn/models/zhaode/Qwen2.5-Math-7B-Instruct-MNN/files
-[modelscope-reader-lm-0.5b]: https://modelscope.cn/models/zhaode/reader-lm-0.5b-MNN/files
-[modelscope-reader-lm-1.5b]: https://modelscope.cn/models/zhaode/reader-lm-1.5b-MNN/files
-
-[modelscope-chatglm-6b]: https://modelscope.cn/models/zhaode/chatglm-6b-MNN/files
-[modelscope-chatglm2-6b]: https://modelscope.cn/models/zhaode/chatglm2-6b-MNN/files
-[modelscope-codegeex2-6b]: https://modelscope.cn/models/zhaode/codegeex2-6b-MNN/files
-[modelscope-chatglm3-6b]: https://modelscope.cn/models/zhaode/chatglm3-6b-MNN/files
-[modelscope-glm4-9b-chat]: https://modelscope.cn/models/zhaode/glm-4-9b-chat-MNN/files
-
-[modelscope-llama2-7b-chat]: https://modelscope.cn/models/zhaode/Llama-2-7b-chat-MNN/files
-[modelscope-llama3-8b-instruct]: https://modelscope.cn/models/zhaode/Llama-3-8B-Instruct-MNN/files
-[modelscope-llama3.2-1b-instruct]: https://modelscope.cn/models/zhaode/Llama-3.2-1B-Instruct-MNN/files
-[modelscope-llama3.2-3b-instruct]: https://modelscope.cn/models/zhaode/Llama-3.2-3B-Instruct-MNN/files
-[modelscope-baichuan2-7b-chat]: https://modelscope.cn/models/zhaode/Baichuan2-7B-Chat-MNN/files
-[modelscope-internlm-chat-7b]: https://modelscope.cn/models/zhaode/internlm-chat-7b-MNN/files
-[modelscope-yi-6b-chat]: https://modelscope.cn/models/zhaode/Yi-6B-Chat-MNN/files
-[modelscope-deepseek-7b-chat]: https://modelscope.cn/models/zhaode/deepseek-llm-7b-chat-MNN/files
-[modelscope-tinyllama-1.1b-chat]: https://modelscope.cn/models/zhaode/TinyLlama-1.1B-Chat-MNN/files
-[modelscope-phi-2]: https://modelscope.cn/models/zhaode/phi-2-MNN/files
-
-[modelscope-bge-large-zh]: https://modelscope.cn/models/zhaode/bge-large-zh-MNN/files
-[modelscope-gte_sentence-embedding_multilingual-base]: https://modelscope.cn/models/zhaode/gte_sentence-embedding_multilingual-base-MNN/files
+[model download](./docs/download.md)
 
 ## Building
 
diff --git a/docs/download.md b/docs/download.md
index d4da3e6..6b279f7 100644
--- a/docs/download.md
+++ b/docs/download.md
@@ -45,4 +45,8 @@
 | [reader-lm-0.5b](https://huggingface.co/jinaai/reader-lm-0.5b) | [Q4_1](https://modelscope.cn/models/MNN/reader-lm-0.5b-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/reader-lm-0.5b-MNN) |
 | [reader-lm-1.5b](https://huggingface.co/jinaai/reader-lm-1.5b) | [Q4_1](https://modelscope.cn/models/MNN/reader-lm-1.5b-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/reader-lm-1.5b-MNN) |
 | [TinyLlama-1.1B-Chat-v1.0](https://modelscope.cn/models/AI-ModelScope/TinyLlama-1.1B-Chat-v1.0/summary) | [Q4_1](https://modelscope.cn/models/MNN/TinyLlama-1.1B-Chat-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/TinyLlama-1.1B-Chat-MNN) |
-| [Yi-6B-Chat](https://modelscope.cn/models/01ai/Yi-6B-Chat/summary) | [Q4_1](https://modelscope.cn/models/MNN/Yi-6B-Chat-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/Yi-6B-Chat-MNN) |
\ No newline at end of file
+| [Yi-6B-Chat](https://modelscope.cn/models/01ai/Yi-6B-Chat/summary) | [Q4_1](https://modelscope.cn/models/MNN/Yi-6B-Chat-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/Yi-6B-Chat-MNN) |
+| [MobileLLM-125M](https://huggingface.co/facebook/MobileLLM-125M) | [Q4_1](https://modelscope.cn/models/MNN/MobileLLM-125M-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/MobileLLM-125M-MNN) |
+| [MobileLLM-350M](https://huggingface.co/facebook/MobileLLM-350M) | [Q4_1](https://modelscope.cn/models/MNN/MobileLLM-350M-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/MobileLLM-350M-MNN) |
+| [MobileLLM-600M](https://huggingface.co/facebook/MobileLLM-600M) | [Q4_1](https://modelscope.cn/models/MNN/MobileLLM-600M-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/MobileLLM-600M-MNN) |
+| [MobileLLM-1B](https://huggingface.co/facebook/MobileLLM-1B) | [Q4_1](https://modelscope.cn/models/MNN/MobileLLM-1B-MNN) | [Q4_1](https://huggingface.co/alibaba-mnn/MobileLLM-1B-MNN) |
\ No newline at end of file