Skip to content

Commit 8403262

Browse files
11zhouxuan11zhouxuan
committed
fix:jina embedding v4
2 parents dd8edb5 + 1ff5900 commit 8403262

34 files changed

+1600
-608
lines changed

‎README.md

Lines changed: 27 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,10 @@
11
<p align="center">
2-
<h3 align="center">Easy Model Deployer: Simple, Efficient, and Easy-to-Integrate</h3>
2+
<h3 align="center">Easy Model Deployer: Easy Foundation Model Hosting on AWS</h3>
3+
</p>
4+
5+
<p align="center">
6+
<a href="README.md"><strong>English</strong></a> |
7+
<a href="README_zh.md"><strong>简体中文</strong></a>
38
</p>
49

510
<p align="center">
@@ -10,21 +15,18 @@
1015
<p align="center">
1116
<a href="https://opensource.org/licenses/MIT"><img src="https://img.shields.io/badge/License-MIT-yellowgreen.svg" alt="MIT License"></a>
1217
<a href="https://pypi.org/project/easy_model_deployer"><img src="https://img.shields.io/pypi/v/easy_model_deployer.svg?logo=pypi&label=PyPI&logoColor=gold"></a>
13-
<a href="https://pypi.org/project/easy_model_deployer"><img alt="PyPI - Downloads" src="https://img.shields.io/pypi/dw/easy-model-deployer"></a>
18+
<a href="https://pypi.org/project/easy_model_deployer"><img alt="PyPI - Downloads" src="https://img.shields.io/pypi/dw/easy_model_deployer"></a>
1419
<a href="https://github.com/aws-samples/easy-model-deployer/actions/workflows/release-package.yml"><img src="https://github.com/aws-samples/easy-model-deployer/actions/workflows/release-package.yml/badge.svg" alt="Build Status"></a>
1520
</p>
1621

17-
## 🔥 Latest News
18-
- 2025-04-29: Deploy Qwen 3 series models with [one command line](https://github.com/aws-samples/easy-model-deployer/blob/main/docs/en/best_deployment_practices.md##famous-models###Qwen-3-Series).
19-
- 2025-04-21: Deploy GLM Z1/0414 series models with [one command line](https://github.com/aws-samples/easy-model-deployer/blob/main/docs/en/best_deployment_practices.md##famous-models###GLM-Z1/0414-Series).
20-
- 2025-03-17: Deploy Gemma 3 series models with [one command line](https://github.com/aws-samples/easy-model-deployer/blob/main/docs/en/best_deployment_practices.md##famous-models###gemma-3-series).
21-
- 2025-03-06: Deploy QwQ-32B with [one command line](docs/en/best_deployment_practices.md##famous-models###qwen-series###qwq-32b).
22+
![header](docs/images/header.jpg)
23+
2224

2325
## Introduction
2426

2527
Easy Model Deployer is a lightweight tool designed for simplify deploy **Open-Source LLMs** ([Supported Models](docs/en/supported_models.md)) and Custom Models on AWS. It provides **OpenAI's Completions API** and [**LangChain Interface**](https://github.com/langchain-ai/langchain). Built for developers who need reliable and scalable model serving without complex environment setup.
2628

27-
![cli](docs/images/demo.avif)
29+
![deploy](docs/images/demo-deploy.avif)
2830

2931
**Key Features**
3032

@@ -34,12 +36,18 @@ Easy Model Deployer is a lightweight tool designed for simplify deploy **Open-So
3436
- Different instance types (CPU/GPU/AWS Inferentia)
3537
- Convenient integration (OpenAI Compatible API, LangChain client, etc.)
3638

37-
## Support Models
38-
<details>
39-
<summary>Deepseek Reasoning Model</summary>
40-
<a href="https://github.com/deepseek-ai/DeepSeek-R1"><strong>DeepSeek-R1-Distill-Qwen-14B</strong></a>
39+
## Supported Models
4140

42-
</details>
41+
Easy Model Deployer supports a wide range of models including:
42+
43+
- **LLMs**: Qwen, Llama, DeepSeek, GLM, InternLM, Baichuan, and more
44+
- **Vision-Language Models**: Qwen-VL, InternVL, Gemma3-Vision, and more
45+
- **Embedding Models**: BGE, Jina, BERT-based models
46+
- **Reranking Models**: BGE-Reranker, Jina-Reranker
47+
- **ASR Models**: Whisper variants
48+
- **Custom Models**: Support for custom Docker images
49+
50+
For the complete list of supported models and deployment configurations, see [Supported Models](docs/en/supported_models.md).
4351

4452
## 🔧 Get Started
4553

@@ -49,6 +57,7 @@ Install Easy Model Deployer with PyPI, currently support for Python 3.9 or above
4957

5058
```bash
5159
pip install easy-model-deployer
60+
5261
emd
5362
```
5463

@@ -93,10 +102,10 @@ emd status
93102
Invoke the deployed model for testing by CLI.
94103

95104
```bash
96-
emd invoke DeepSeek-R1-Distill-Qwen-1.5B
105+
emd invoke <ModelId>
97106
```
98107

99-
> **💡 Tip** You can find the *ModelId* in the output by `emd status`.
108+
> **💡 Tip** You can find the *ModelId* in the output by `emd status`. For example: `emd invoke DeepSeek-R1-Distill-Qwen-1.5B`
100109
101110
- [Integration examples](https://aws-samples.github.io/easy-model-deployer/)
102111
- [EMD client](docs/en/emd_client.md)
@@ -107,7 +116,7 @@ emd invoke DeepSeek-R1-Distill-Qwen-1.5B
107116
108117
### List Supported Models
109118

110-
Quickly see what models are supported, this command will output all information related to deployment. (Plese browse [Supported Models](docs/en/supported_models.md) for more information.)
119+
Quickly see what models are supported, this command will output all information related to deployment. (Please browse [Supported Models](docs/en/supported_models.md) for more information.)
111120

112121
```bash
113122
emd list-supported-models
@@ -119,10 +128,10 @@ emd list-supported-models
119128
Delete the deployed model.
120129

121130
```bash
122-
emd destroy DeepSeek-R1-Distill-Qwen-1.5B
131+
emd destroy <ModelId>
123132
```
124133

125-
> **💡 Tip** You can find the *ModelId* in the output by `emd status`.
134+
> **💡 Tip** You can find the *ModelId* in the output by `emd status`. For example: `emd destroy DeepSeek-R1-Distill-Qwen-1.5B`
126135
127136
## 📖 Documentation
128137

‎README_zh.md

Lines changed: 144 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,144 @@
1+
<p align="center">
2+
<h3 align="center">Easy Model Deployer: 大模型上云，一键搞定</h3>
3+
</p>
4+
5+
<p align="center">
6+
<a href="README.md"><strong>English</strong></a> |
7+
<a href="README_zh.md"><strong>简体中文</strong></a>
8+
</p>
9+
10+
<p align="center">
11+
<a href="https://aws-samples.github.io/easy-model-deployer/en/installation"><strong>文档</strong></a> ·
12+
<a href="https://github.com/aws-samples/easy-model-deployer/releases"><strong>更新日志</strong></a>
13+
</p>
14+
15+
<p align="center">
16+
<a href="https://opensource.org/licenses/MIT"><img src="https://img.shields.io/badge/License-MIT-yellowgreen.svg" alt="MIT License"></a>
17+
<a href="https://pypi.org/project/easy_model_deployer"><img src="https://img.shields.io/pypi/v/easy_model_deployer.svg?logo=pypi&label=PyPI&logoColor=gold"></a>
18+
<a href="https://pypi.org/project/easy_model_deployer"><img alt="PyPI - Downloads" src="https://img.shields.io/pypi/dw/easy_model_deployer"></a>
19+
<a href="https://github.com/aws-samples/easy-model-deployer/actions/workflows/release-package.yml"><img src="https://github.com/aws-samples/easy-model-deployer/actions/workflows/release-package.yml/badge.svg" alt="Build Status"></a>
20+
</p>
21+
22+
![header](docs/images/header.jpg)
23+
24+
## 简介
25+
26+
还在为大模型部署而头疼吗？Easy Model Deployer 让你告别复杂的环境配置，轻松将**开源大模型**（[支持的模型](docs/en/supported_models.md)）部署到 AWS 云端。
27+
28+
无论是大语言模型、视觉模型还是自定义模型，一条命令即可搞定部署。更棒的是，部署完成后直接获得 **OpenAI 兼容 API** 和 [**LangChain 接口**](https://github.com/langchain-ai/langchain)，让你的 AI 应用开发如虎添翼。
29+
30+
专为追求高效、稳定模型服务的开发者打造，让技术门槛不再成为创新的阻碍。
31+
32+
![deploy](docs/images/demo-deploy.avif)
33+
34+
**核心特性**
35+
36+
- 一键部署模型到 AWS（Amazon SageMaker、Amazon ECS、Amazon EC2）
37+
- 多样化模型类型（大语言模型、视觉语言模型、嵌入模型、视觉模型等）
38+
- 丰富的推理引擎（vLLM、TGI、Lmdeploy 等）
39+
- 不同实例类型（CPU/GPU/AWS Inferentia）
40+
- 便捷集成（OpenAI 兼容 API、LangChain 客户端等）
41+
42+
## 支持的模型
43+
44+
Easy Model Deployer 支持广泛的模型类型，包括：
45+
46+
- **大语言模型**: Qwen、Llama、DeepSeek、GLM、InternLM、Baichuan 等
47+
- **视觉语言模型**: Qwen-VL、InternVL、Gemma3-Vision 等
48+
- **嵌入模型**: BGE、Jina、基于 BERT 的模型
49+
- **重排序模型**: BGE-Reranker、Jina-Reranker
50+
- **语音识别模型**: Whisper 变体
51+
- **自定义模型**: 支持自定义 Docker 镜像
52+
53+
完整的支持模型列表和部署配置，请参见[支持的模型](docs/en/supported_models.md)。
54+
55+
## 🔧 快速开始
56+
57+
### 安装
58+
59+
通过 PyPI 安装 Easy Model Deployer，目前支持 Python 3.9 及以上版本：
60+
61+
```bash
62+
pip install easy-model-deployer
63+
64+
emd
65+
```
66+
67+
### 初始化
68+
69+
准备模型部署所需的基础资源。
70+
71+
更多信息请参考[架构说明](https://aws-samples.github.io/easy-model-deployer/en/architecture/)。
72+
73+
```bash
74+
emd bootstrap
75+
```
76+
77+
> **💡 提示** 通过 `pip` 升级 EMD 后，需要重新运行此命令来更新环境。
78+
79+
### 部署模型
80+
81+
通过交互式 CLI 或一行命令部署模型。
82+
83+
```bash
84+
emd deploy
85+
```
86+
87+
> **💡 提示** 要查看所有可用参数，请运行 `emd deploy --help`。
88+
> 当您看到 "Waiting for model: ..." 消息时，表示部署任务已开始，您可以按 `Ctrl+C` 停止终端输出。
89+
>
90+
> - 有关部署参数的更多信息，请参考[部署参数](docs/en/installation.md)。
91+
> - 有关使用命令行参数的最佳实践示例，请参考[最佳部署实践](docs/en/best_deployment_practices.md)。
92+
93+
### 查看状态
94+
95+
检查模型部署任务的状态。
96+
97+
```bash
98+
emd status
99+
```
100+
101+
> **💡 提示** EMD 允许同时启动多个部署任务。
102+
103+
### 调用模型
104+
105+
通过 CLI 调用已部署的模型进行测试。
106+
107+
```bash
108+
emd invoke <ModelId>
109+
```
110+
111+
> **💡 提示** 您可以在 `emd status` 的输出中找到 *ModelId*。例如：`emd invoke DeepSeek-R1-Distill-Qwen-1.5B`
112+
113+
- [集成示例](https://aws-samples.github.io/easy-model-deployer/)
114+
- [EMD 客户端](docs/en/emd_client.md)
115+
- [Langchain 接口](docs/en/langchain_interface.md)
116+
- [OpenAI 兼容接口](docs/en/openai_compatiable.md)
117+
118+
> **💡 提示** OpenAI 兼容 API 仅支持 Amazon ECS 和 Amazon EC2 部署类型。
119+
120+
### 列出支持的模型
121+
122+
快速查看支持哪些模型，此命令将输出与部署相关的所有信息。（更多信息请浏览[支持的模型](docs/en/supported_models.md)。）
123+
124+
```bash
125+
emd list-supported-models
126+
```
127+
128+
### 删除模型
129+
130+
删除已部署的模型。
131+
132+
```bash
133+
emd destroy <ModelId>
134+
```
135+
136+
> **💡 提示** 您可以在 `emd status` 的输出中找到 *ModelId*。例如：`emd destroy DeepSeek-R1-Distill-Qwen-1.5B`
137+
138+
## 📖 文档
139+
140+
有关高级配置和详细指南，请访问我们的[文档网站](https://aws-samples.github.io/easy-model-deployer/)。
141+
142+
## 🤝 贡献
143+
144+
我们欢迎贡献！请查看 [CONTRIBUTING.md](CONTRIBUTING.md) 了解指南。

‎docs/en/installation.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -64,7 +64,7 @@ uv pip install easy-model-deployer
6464
**Verification:**
6565
After installation, verify that EMD is working correctly by running:
6666
```bash
67-
emd --version
67+
emd version
6868
```
6969
This should display the installed version of Easy Model Deployer.
7070

0 commit comments

Comments
 (0)