From 5136ecb9fe4e4753bfc6633884f711717f17466d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=88=98=E4=B8=B9?= <liudan@MacBook-Pro.local>
Date: Wed, 5 Jun 2024 18:05:24 +0800
Subject: [PATCH] add ollama support minicpm-1b

---
 README.md | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/README.md b/README.md
index 30f26f6..dbd5324 100644
--- a/README.md
+++ b/README.md
@@ -204,11 +204,34 @@ MiniCPM支持[llama.cpp](https://github.com/ggerganov/llama.cpp/) 、[ollama](ht
 更多参数调整[详见](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
 
 **ollama**
+
+***ollama自动安装模型***
 1. [安装ollama](https://github.com/ollama/ollama)
 2. 在命令行运行:
 ```
 ollama run modelbest/minicpm-2b-dpo
 ```
+***ollama手动安装模型***
+1. [安装ollama](https://github.com/ollama/ollama)
+2. 下载gguf形式的模型。[下载链接2b-fp16格式](https://huggingface.co/runfuture/MiniCPM-2B-dpo-fp16-gguf) [下载链接2b-q4km格式](https://huggingface.co/runfuture/MiniCPM-2B-dpo-q4km-gguf)[下载链接1b-fp16格式](https://huggingface.co/linglingdan/MiniCPM-1b-fp16-gguf)[下载链接1b-qr_1格式](https://huggingface.co/linglingdan/MiniCPM-1b-q4-1)
+3. 在命令行运行以下命令,model_name可自定义：
+```
+touch model_name.Modelfile
+```
+4. 将以上model_name.Modelfile的内容修改如下,FROM空格后写入gguf的模型路径
+```
+FROM model_path/model_name.gguf
+TEMPLATE """<s><USER>{{ .Prompt }}<AI>{{ .Response }}"""
+PARAMETER stop "<\s>"
+```
+5. 在命令行运行以下命令，创建ollama模型，ollama_model_name可自定义，model_name.Modelfile参考第3步命名
+```
+ollama create ollama_model_name -f model_name.Modelfile
+```
+6. 运行ollama模型：
+```
+ollama run ollama_model_name
+```
 
 **fastllm**
 1. [编译安装fastllm](https://github.com/ztxz16/fastllm)