diff --git a/README.md b/README.md
new file mode 100644
index 0000000..25dba6e
--- /dev/null
+++ b/README.md
@@ -0,0 +1,113 @@
+# Gorm Sharding
+
+Gorm Plugin for sharding table.
+
+This plugin base on SQL Parser, catch raw SQL before execute db, and then replace table name by sharding rules.
+
+## Translations
+
+For chinese, see [README_zh.md](./README_zh.md)
+
+## Features
+
+- Non-intrusive design. Load the plugin, specify the config, and all done.
+- Lighting-fast. No network based middlewares, as fast as Go.
+- Multiple database support. PostgreSQL tested, MySQL and SQLite is coming.
+
+## Install
+
+```bash
+go get -u github.com/longbridgeapp/gorm-sharding
+```
+
+## Usage
+
+After the database connection opened, use the sharding plugin that registered the tables which you want to shard.
+
+The `Register` function takes a map, the key is the **original table name** and the value is a **resolver** which is composed by five configurable fields that described in [Config description](config-description).
+
+## Usage Example
+
+```go
+sharding := gormsharding.Register(map[string]gormsharding.Resolver{
+	"orders": {
+		EnableFullTable: true,
+		ShardingColumn: "user_id",
+		ShardingAlgorithm: func(value interface{}) (suffix string, err error) {
+			switch user_id := value.(type) {
+			case int64:
+				return fmt.Sprintf("_%02d", user_id % 64), nil
+			default:
+				return "", errors.New("invalid user_id")
+			}
+		},
+		ShardingAlgorithmByPrimaryKey: func(id int64) (suffix string) {
+			return fmt.Sprintf("_%02d", keygen.TableIdx(id))
+		},
+		PrimaryKeyGenerate: func(tableIdx int64) int64 {
+			return keygen.Next(tableIdx)
+		}
+	},
+})
+db.Use(sharding)
+```
+
+## Config description
+
+- [EnableFullTable](#EnableFullTable)
+- [ShardingColumn](#ShardingColumn)
+- [ShardingAlgorithm](#ShardingAlgorithm)
+- [ShardingAlgorithmByPrimaryKey](#ShardingAlgorithmByPrimaryKey)
+- [PrimaryKeyGenerate](#PrimaryKeyGenerate)
+
+#### EnableFullTable
+
+Whether to enable a full table.
+
+When full table enabled, sharding plugin will double write data to both main table and sharding table.
+
+#### ShardingColumn
+
+Which table column you want to used for sharding the table rows.
+
+For example, for a product order table, you may want to split the rows by `user_id`.
+
+### ShardingAlgorithm
+
+A function to generate the sharding table's suffix by the column value.
+
+It's signature is `func(columnValue interface{}) (suffix string, err error)`.
+
+For an example, see the [usage example](#usage-example) above.
+
+#### ShardingAlgorithmByPrimaryKey
+
+A function to generate the sharding table's suffix by the primary key.
+
+It's signature is `func(id int64) (suffix string)`.
+
+For an example, see the [usage example](#usage-example) above.
+
+Note, when the record contains an id field, ShardingAlgorithmByPrimaryKey will preferred than ShardingAlgorithm.
+
+#### PrimaryKeyGenerate
+
+A function to generate the primary key.
+
+Used only when insert and the record does not contains an id field.
+
+It's signature is `func(tableIdx int64) int64`.
+
+For an example, see the [usage example](#usage-example) above.
+
+We recommend you use the [keygen](https://github.com/longbridgeapp/gorm-sharding) component, it is a distributed primary key generator.
+
+When use auto-increment like generator, the tableIdx argument could ignored.
+
+## References
+
+https://gorm.io/docs/write_plugins.html#Plugin
+
+## License
+
+This project under MIT license.
diff --git a/README_zh.md b/README_zh.md
new file mode 100644
index 0000000..2a6fbd1
--- /dev/null
+++ b/README_zh.md
@@ -0,0 +1,110 @@
+# Gorm Sharding
+
+Gorm 分表插件。
+
+基于 SQL 解析器，执行前捕获原始 SQL，根据分表规则替换表名。
+
+## 特色
+
+- 非侵入式设计。加载插件，指定配置，完成。
+- 快。没有网络中间件，跟 Go 一样快。
+- 多数据库支持。PostgreSQL 已测试，MySQL 和 SQLite 进行中。
+
+## 安装
+
+```bash
+go get -u github.com/longbridgeapp/gorm-sharding
+```
+
+## 用法
+
+数据库连接打开后，使用分表插件注册需要拆分的表。
+
+`Register` 函数接收一个 map，键是**原始表名**，值是一个 **resolver** ，由可配置的字段组成，见配置描述。
+
+## 用法示例
+
+```go
+sharding := gormsharding.Register(map[string]gormsharding.Resolver{
+	"orders": {
+		EnableFullTable: true,
+		ShardingColumn: "user_id",
+		ShardingAlgorithm: func(value interface{}) (suffix string, err error) {
+			switch user_id := value.(type) {
+			case int64:
+				return fmt.Sprintf("_%02d", user_id % 64), nil
+			default:
+				return "", errors.New("invalid user_id")
+			}
+		},
+		ShardingAlgorithmByPrimaryKey: func(id int64) (suffix string) {
+			return fmt.Sprintf("_%02d", keygen.TableIdx(id))
+		},
+		PrimaryKeyGenerate: func(tableIdx int64) int64 {
+			return keygen.Next(tableIdx)
+		}
+	},
+})
+db.Use(sharding)
+```
+
+## 配置描述
+
+- [EnableFullTable](#EnableFullTable)
+- [ShardingColumn](#ShardingColumn)
+- [ShardingAlgorithm](#ShardingAlgorithm)
+- [ShardingAlgorithmByPrimaryKey](#ShardingAlgorithmByPrimaryKey)
+- [PrimaryKeyGenerate](#PrimaryKeyGenerate)
+
+#### EnableFullTable
+
+Whether to enable a full table.
+是否开启全表。
+
+开启后，分表插件将双写数据到主表和分表。
+
+#### ShardingColumn
+
+用来分表的表字段。
+
+如，对于一个订单表，你可能想用 `user_id` 拆分数据。
+
+### ShardingAlgorithm
+
+使用列值生成分表后缀的函数。
+
+签名是 `func(columnValue interface{}) (suffix string, err error)`。
+
+可以参考上面的用法示例。
+
+#### ShardingAlgorithmByPrimaryKey
+
+使用主键生成分表后缀的函数。
+
+签名是 `func(id int64) (suffix string)`。
+
+可以参考上面的用法示例。
+
+注意，当记录包含 id 字段时，ShardingAlgorithmByPrimaryKey 优先于 ShardingAlgorithm。
+
+#### PrimaryKeyGenerate
+
+生成主键的函数。
+
+只有当插入数据并且记录不包含 id 字段时使用。
+
+签名是 `func(tableIdx int64) int64`。
+
+可以参考上面的用法示例。
+
+推荐使用 [keygen](https://github.com/longbridgeapp/gorm-sharding) 组件，它是一个分布式的主键生成器。
+
+当使用自增类的生成器时，tableIdx 参数可以忽略。
+
+## 参考
+
+https://gorm.io/docs/write_plugins.html#Plugin
+
+## 许可证
+
+本项目使用 MIT 许可证。