DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Xl Free Model — DSH Plugin for DeepSeek Harness
← Plugins
X

dsh-xl-free-model

Xl Free Model

XL Free Model — Install this plugin in dsh to automatically run a local OpenAI-compatible proxy and use free models such as mimo-v2.6-flash-free through the OpenCode free channel; the models are registered directly in the session model selection list (provider: xl-free-model), with no need to manual

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add github:zhangmaorongXL/dsh-xl-free-model#983e871008dd24440213557682a752a223e7839e
READMECompatibilityVersions

Description

XL Free Model — Install this plugin in dsh to automatically run a local OpenAI-compatible proxy and use free models such as mimo-v2.6-flash-free through the OpenCode free channel; the models are registered directly in the session model selection list (provider: xl-free-model), with no need to manually configure the provider in the profile; the settings page includes an XL Free Model tab displaying the proxy status; supports streaming, multi-turn session reuse, and function calling bridging.

Compatibility and provenance

Xl Free Model is published as dsh-xl-free-model and currently resolves to version 1.4.2. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
web
Release source
github
Registry updated
10/15/2026

Versions

1.4.2stable
10/15/2026
1.4.1stable
10/15/2026

Related plugins

Loading related plugins…

Latest
1.4.2
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
web
License
MIT
Source
github
GitHub
★ 0
Weekly downloads
0
Last push
10/15/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in

Related plugins

More verified plugins in models-usage.

Whale Widgetdsh-whale-widgetDeepSeek balance whale widget in the bottom-right corner of the DSH Web interface: balance/today’s usage/peak-off-peak pricing, customizable bubble click sequence (text/balance/today/peak-off-peak/image/random phrases and parallel weighted selection), per-line styles and fonts, floating quick editinUsage@linxin666/dsh-usageUsage statistics plugin for the dsh web GUI: per-provider balance and coding-plan quota detection plus a live token usage ledger, with a compact price glance card above the sidebar's Settings rowClient Ui Model Capabilities@linxin666/dsh-client-ui-model-capabilitiesPer-model capability declarations (image input and reasoning efforts) for custom pi-ai providers, edited in place on the Models settings cards; writes the official llm-pi-ai settings namespace, no DSH source changes.Cost Meterdsh-cost-meterDeepSeek Harness session cost statistics plugin: current session cost, today’s charges, historical records, and synchronization with official pricing; supports multi-provider, multi-model price calculation (built-in pricing catalog for 170+ model IDs with automatic matching); queries and displays su

README

dsh-xl-free-model

在 dsh 中自动运行本地 OpenAI 兼容代理的插件,经 OpenCode 免费通道使用 mimo-v2.6-flash-free 等 free 模型。支持流式、多轮会话复用、function calling 桥接与无头 CLI 后备(桌面版未打开时自动拉起)。

模型通过 Host 半身注册的 LlmAdapter 直接出现在会话的模型选择列表中(provider: xl-free-model),无需在 profile 的 cordis.patch.yml 手工配置 provider。设置页含 XL Free Model 插页,展示代理状态与模型卡片列表。

功能特性

  • 零配置模型接入:安装插件后,免费模型自动出现在会话的模型选择列表
  • 本地代理:OpenAI 兼容 API(http://127.0.0.1:8787/v1),可被任意 OpenAI SDK 调用
  • 自动管理:dsh 启动时自动拉起代理,退出时自动清理
  • 桌面/无头双后端:优先复用已打开的 OpenCode 桌面版;未打开时自动启动无头 CLI 后备
  • CLI 二进制自动安装:缺失时直接从 npm registry 下载 tarball 解压(无需 npm 在 PATH),npm install 兜底
  • 流式支持:完整支持 SSE 流式响应
  • Function Calling:通过提示词桥接模拟工具调用
  • 会话复用:后端连接池自动复用,提升响应速度
  • PEB 密码发现:无需手动配置密码,自动从 OpenCode 进程环境读取

安装

前置条件

  • Windows 10/11 x64
  • Node.js ≥ 22
  • dsh 桌面版或 CLI

步骤

  1. 克隆仓库

    git clone https://github.com/zhangmaorongXL/dsh-xl-free-model.git
    cd dsh-xl-free-model
    
  2. 安装到 dsh

    将本目录复制到 dsh 插件目录,或在 dsh 设置中指定插件路径。

  3. 重启 dsh

    重启后,会话模型选择列表中将出现 XL Free Model 分组,包含可用的免费模型。

使用

在 dsh 中使用

  1. 新建会话
  2. 在模型选择器中选择 XL Free Model 分组下的模型(如 mimo-v2.6-flash-free)
  3. 正常对话即可

独立使用代理

代理兼容 OpenAI API,可被任意支持自定义 base URL 的工具调用:

# 启动代理(dsh 运行时会自动管理,通常无需手动)
node proxy.cjs
# Python 示例
from openai import OpenAI

client = OpenAI(
    base_url="http://127.0.0.1:8787/v1",
    api_key="changeme-dsh-key"  # proxy.config.json 中的 apiKey
)

response = client.chat.completions.create(
    model="mimo-v2.6-flash-free",
    messages=[{"role": "user", "content": "你好"}],
    stream=True
)
// Node.js 示例
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "http://127.0.0.1:8787/v1",
  apiKey: "changeme-dsh-key",
});

const stream = await client.chat.completions.create({
  model: "mimo-v2.6-flash-free",
  messages: [{ role: "user", content: "你好" }],
  stream: true,
});

配置

proxy.config.json 完整配置项:

{
  "proxyHost": "127.0.0.1",        // 代理监听地址
  "proxyPort": 8787,               // 代理监听端口
  "apiKey": "changeme-dsh-key",    // API Key(必改)
  "backend": {
    "host": "",                    // OpenCode 后端地址(留空自动发现)
    "port": 0,                     // OpenCode 后端端口(留空自动发现)
    "username": "opencode",        // 后端用户名
    "password": "",                // 后端密码(留空自动 PEB 发现)
    "cli": {
      "enabled": true,             // 是否启用无头 CLI 后备
      "path": "cli\\node_modules\\opencode-windows-x64\\bin\\opencode.exe",
      "port": 4097                 // CLI 后备端口
    }
  }
}

架构

dsh-xl-free-model/
├── index.js              # Host 半身:挂载代理 + 注册 LlmAdapter + CLI 二进制自动安装
├── adapter.js            # LlmAdapter 实现(dsh 结构化契约)
├── client.js             # Browser 半身:设置页插页 UI
├── proxy.cjs             # 本地 OpenAI 兼容代理服务器
├── peb_env.ps1           # PEB 密码发现脚本(PowerShell)
├── proxy.config.json     # 代理配置
├── cordis.patch.yml      # dsh bundle patch(仅挂载 Host)
├── package.json          # 插件清单
├── cli/                  # OpenCode 无头 CLI 后备
│   ├── package.json      # npm 依赖清单(opencode-ai)
│   └── node_modules/     # 自动下载/npm install 后生成(.gitignore 排除)
├── .gitignore            # 排除大文件与运行时产物
├── LICENSE               # MIT
└── README.md

工作流程

  1. dsh 加载插件 → index.js 的 apply() 执行
  2. 读取 proxy.config.json,注入 API Key 环境变量
  3. 检测代理是否已在运行;未运行则 spawn proxy.cjs
  4. 通过 ctx.llm.registerAdapter() 注册 xl-free-model provider
  5. adapter.js 的 listModels() 从代理拉取模型列表
  6. 会话选择模型时,adapter.js 的 stream() 转发请求到代理
  7. 代理转发到 OpenCode 后端(桌面版或无头 CLI),返回流式响应

故障排查

模型未出现在选择列表

  • 检查 dsh 日志(~/.dsh/logs/)中 registerAdapter failed 错误
  • 确认代理已启动:curl http://127.0.0.1:8787/health
  • 重启 dsh

代理启动失败

  • 检查端口占用:netstat -ano | findstr 8787
  • 查看 cli-serve.log(CLI 后备日志)
  • 确认 cli/node_modules/opencode-windows-x64/bin/opencode.exe 存在

密码获取失败

  • 确认 OpenCode 桌面版已打开
  • 手动测试:powershell -NoProfile -ExecutionPolicy Bypass -File peb_env.ps1 OPENCODE_SERVER_PASSWORD
  • 或在 proxy.config.json 的 backend.password 中手动填写

开发

本地测试

# 语法检查
node --check index.js
node --check adapter.js
node --check client.js
node --check proxy.cjs

# 启动代理(前台运行)
node proxy.cjs

修改适配器

adapter.js 实现了 dsh 的结构化 LlmAdapter 契约:

  • providerInfo(provider) → { id, name }
  • listModels(provider) → LlmModelInfo[]
  • resolveModel(provider, model) → ResolvedModelInfo
  • prepareCall(provider, model) → { model, stream }
  • stream(options) → AsyncIterable<StreamChunk>

参考 dsh 源码中的 LlmAdapter 类型定义。

许可证

MIT