Certainly! Here's how you can translate "Demo Open Web UI with Models" into simplified Chinese, while keeping the HTML structure intact: ```html 演示使用模型开放Web UI ``` This HTML snippet ensures that the text "演示使用模型开放Web UI" is displayed in simplified Chinese.

Here’s the translation while keeping the HTML structure: ```html A guide to host your first LLM-based chat service in Docker 在Docker中托管您的第一个基于LLM的聊天服务指南 ```

Photo by Mariia Shalabaieva on Unsplash

Certainly! Here's how you can structure the HTML with the translated text in simplified Chinese: ```html

Translate to Chinese

最近在生成AI技术的炒作下,许多组织都对基于大语言模型(LLM)的PoC聊天服务在其产品线上表现出浓厚兴趣。

作为基础架构工程师,我花了一些时间部署了一个开放AI服务。我还在我的GitHub仓库中创建了一个演示项目chat-service,其中包括了一份逐步部署指南。

该项目支持两种模型来源,以展示其灵活性。此外,我写这篇文章来讨论演示中的考虑因素。

``` In this HTML structure: - The `

` tag is used for the main heading. - Two `

` tags are used for paragraphs of text. - The content inside the tags is the translated text in simplified Chinese.

Sure, here's how you would translate "Architecture" to simplified Chinese, while maintaining the HTML structure: ```html 建筑 ``` In this HTML snippet: - `` is used to encapsulate the text, providing inline styling. - `lang="zh-CN"` specifies the language of the enclosed text as simplified Chinese. - `建筑` is the translation of "Architecture" into simplified Chinese characters. This way, you maintain both the structure and the translation within the HTML context.

Sure, here's the translated text in simplified Chinese, while maintaining the HTML structure: ```html

我们基于几个开源项目进行开发,其中最值得一提的是Open WebUI项目、Ollama项目和LiteLLM项目。Open WebUI项目显然是用户界面。Ollama和LiteLLM负责管理实际的模型。这些模型可以在本地或远程运行。Open WebUI通过标准API与模型管理器通信。此外,我们还需要使用PostgreSQL作为LiteLLM的数据库,以及Nginx作为Open Web UI的反向代理。

``` This HTML structure ensures that the translated text is properly formatted and can be integrated into a webpage if needed.

```html

这里是Ollama支持的模型长列表,我们在这里使用它来本地托管流行的Lama3模型,这需要计算能力。另一方面,LiteLLM作为代理连接到可用的模型。例如,Open AI平台运行一些模型,如GPT-4o、GPT-3.5 Turbo等。Azure Open AI服务也托管类似的一组模型。在本教程中,我们使用来自Open AI平台的GPT-3.5 Turbo。您需要在Open AI平台创建一个付费账户,并提供一个API密钥。

```
Components of the chat service

Sure, here's the translated text in simplified Chinese while keeping the HTML structure intact: ```html

在同一台服务器上有许多组件时,将它们运行在容器中以避免依赖问题是有意义的。幸运的是,它们都发布了自己的容器镜像,我们可以使用 Docker Compose 来编排它们,这是这个演示项目的主要成果。

``` This HTML snippet now contains the translated text in simplified Chinese.

```html 基础设施 ```

Sure, here's the translated text in simplified Chinese: ```html 所有组件运行在一个服务器上。服务器需要访问互联网以下载容器镜像。测试用户必须能够通过服务器上的TCP端口443访问HTTPS服务。服务器可以是物理的或虚拟的。在我的项目中,我在AWS的EC2实例上通过Docker守护程序运行组件。然而,由于Docker守护程序的跨平台能力,任何虚拟机平台都应该可以使用。 ```

```html

一个特殊的硬件要求是需要GPU,因为我们在本地运行Llama3模型。从技术上讲,我们可以仅使用CPU来运行,但性能会非常慢。如果您没有带GPU的服务器,请考虑使用云服务提供商提供的虚拟GPU运行虚拟机,比如AWS的G4实例和Azure的NV系列。在这个项目中,我选择了AWS的g4dn.xlarge实例类型,配备了NVIDIA T4 GPU、4个虚拟CPU和16GiB内存。关于磁盘空间,为了下载Llama3模型,我们需要大约10GB。我建议根卷只起始配备100GB,因为它们价格相对便宜。在操作系统镜像方面,我选择了当前的Ubuntu镜像。

```

To execute the steps above on AWS in a consistent manner, I recommend using Infrastructure as Code using Terraform. I created the vpc-base repo for this purpose. Please follow the instruction with the chat service use case, and provision the VPC, associated subnets and the EC2 instance in your own AWS account. 保持HTML结构,将上述英文文本翻译成简体中文: 为了在AWS上以一致的方式执行以上步骤,我建议使用Terraform进行基础设施即代码。我为此目的创建了vpc-base仓库。请按照聊天服务用例的说明,在您自己的AWS账户中配置VPC、相关子网和EC2实例。

Sure! Here’s the translation while keeping the HTML structure: ```html

是的,假设你手头没有带 GPU 的服务器来托管 llama3 模型,从云服务提供商那里创建这些资源确实需要很多。不过,如果你只是想使用 LiteLLM 来管理远程模型,可以跳过云基础设施,直接在你的 MacBook 或笔记本电脑上运行容器。

```

Certainly! Here's the text "Containers" translated into simplified Chinese while keeping the HTML structure: ```html 容器 ``` In this translation: - `` is used to inline the text within the HTML structure. - `lang="zh-CN"` specifies the language code for simplified Chinese. - "容器" is the simplified Chinese translation of "Containers".

Sure, here is the HTML structure with the text translated into simplified Chinese: ```html

Translated Text

Docker compose,允许我们定义多个 Docker 容器及其依赖关系,并配置它们之间的通信方式。Docker 在开发者社区中非常知名。我使用它是因为它不依赖于任何特定的云服务提供商,因此具有很好的可移植性。更重要的是,Docker compose 相当简单但足以满足这个演示项目的需求。

``` In this translation: - "Docker compose" is translated as "Docker compose" (with a comma in Chinese). - Key technical terms like "Docker" and "Docker compose" are kept in English within the Chinese text, as they are commonly understood in the Chinese developer community. - The translation conveys the original meaning accurately while respecting the technical context.

Sure, here's the translated text in simplified Chinese, maintaining the HTML structure as requested: ```html

YAML 文件使用了几个常见的组合元素,例如环境变量、卷和端口。其中一个特殊的元素是 Ollama 容器的部署部分:

``` This HTML snippet now contains the translated text in simplified Chinese.
  ollama-svc:
image: ollama/ollama:latest
container_name: ollama
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: 1
capabilities: [gpu]

Certainly! Here's the translation of the provided English text into simplified Chinese while keeping the HTML structure: ```html 通过让容器选择具有 GPU 功能的 nvidia 设备驱动程序,我们允许在其容器中管理 llama3 模型的 Ollama 进程使用由底层操作系统管理的 GPU。还需注意的是,NVIDIA 驱动程序必须先在操作系统上安装。 ```

Sure! Here’s the translated text in simplified Chinese, keeping the HTML structure intact: ```html

您可以在两个配置文件中使用该项目。除了 Nginx 代理和 Open Web UI 服务外,local-model 配置文件包括 ollama 服务,并且必须在具有 NVIDIA GPU 的服务器上运行。remote model 包括 LiteLLM 代理服务,可以在任何普通硬件上运行。您也可以同时运行这两个配置文件。我在 Docker Compose 中使用 profiles 元素来控制基于所选配置文件应启用哪些服务。这使我能够在一个 Compose 文件中管理这两个选项。

```

Sure, here's how you could structure it in HTML while including the simplified Chinese translation: ```html

一旦服务启动,请作为新用户注册并登录。然后我们可以从下拉菜单中选择一个模型进行聊天。我们还可以从用户界面上传文档。

``` This HTML snippet includes the translated text in simplified Chinese, maintaining the basic structure for content presentation.

Certainly! Here is "Operation" translated into simplified Chinese while keeping the HTML structure: ```html 操作 ``` This HTML snippet ensures the text "操作" is displayed in simplified Chinese.

Certainly! Here's the translation of the text into simplified Chinese, while keeping the HTML structure: ```html

显然,这个项目不是为了生产环境。要在生产环境中运行此服务,我们需要考虑更多因素。

``` This HTML snippet contains the translated text in simplified Chinese, maintaining the original structure.

Sure, here's the translated text in simplified Chinese, keeping the HTML structure intact: ```html

为了使服务运行起来,单个服务器是不够的。我们需要一组分布在多个位置的虚拟机,以提供高可用性。我们需要编排大量基于容器的服务的生命周期,这些服务分布在虚拟机农场和 Kubernetes 集群中,Kubernetes 是一个常见的平台选择。

``` Translated text (simplified Chinese): 为了使服务运行起来,单个服务器是不够的。我们需要一组分布在多个位置的虚拟机,以提供高可用性。我们需要编排大量基于容器的服务的生命周期,这些服务分布在虚拟机农场和 Kubernetes 集群中,Kubernetes 是一个常见的平台选择。

```html

在 YAML 文件中避免放置敏感数据(如数据库凭据和 API 密钥)是最佳实践。尽管在这个项目中我们将它们分别放在环境变量文件中,但这仍未达到安全标准。实际解决方案通常涉及专用的密钥存储,具备加密和自动轮换功能。

```

Certainly! Here's the translation in simplified Chinese, keeping the HTML structure intact: ```html

这个项目使用容器运行PostgreSQL服务,但当服务扩展时,由于存储方面的细微差别,可能不太适合在容器平台上托管数据服务。

```

Certainly! Here's the translated text in simplified Chinese, while keeping the HTML structure: ```html 总结 ``` In this HTML snippet, `` is used to encapsulate the translated text "总结" (which means "Summary" in simplified Chinese).

当然!以下是将您的英文文本翻译成简体中文的内容,同时保持了 HTML 结构: ```html

本项目的目标是在 30 分钟内(如果您熟悉技术栈的话)搭建自己的聊天服务。如果您对生成式 AI 的应用有兴趣,这也可以作为在您自己组织内进行概念验证的起点,例如处理内部文档。然而,想要将其扩展到大规模应用仍然需要付出很多努力。

``` 如果您需要对 HTML 结构进行进一步的调整或添加其他内容,请随时告诉我!

2024-07-15 04:27:24 AI中文站翻译自原文