Skip to content

feat: upgrade MiniMax provider to M2.7 with correct API base URL #85

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
octo-patch wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat: upgrade MiniMax provider to M2.7 with correct API base URL #85

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 35 additions & 23 deletions README_zh.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,14 +37,14 @@
---

<a id="核心特性"></a>

## ✨ 核心特性

| | MCP & Skills | 远程控制 | 图形界面操作 |
| ------------- | ------------ | -------------- | ------------- |
| Claude Cowork | ✓ | ✗ | ✗ |
| OpenClaw | ✓ | ✓ | ✗ |
| OpenCowork | ✓ | ✓ | ✓ |

| ------------- | ------------ | -------- | ------------ |
| Claude Cowork | ✓ | ✗ | ✗ |
| OpenClaw | ✓ | ✓ | ✗ |
| OpenCowork | ✓ | ✓ | ✓ |

- **一键安装，开箱即用**：提供 Windows 和 macOS 预构建安装包，无需配置环境，下载即可开始使用。。
- **灵活模型支持**：支持 **Claude**、**OpenAI 兼容接口**，以及国产大模型 **GLM**、**MiniMax**、**Kimi** 等。使用你的 OpenRouter、Anthropic等API Key，灵活配置。更多模型持续接入中！
Expand All @@ -60,22 +60,27 @@
- **UI优化**：灵活优美的UI设计、切换系统语言、完善的MCP/Skills/Tools调用展示。

<a id="演示"></a>

## 🎬 演示

观看 Open Cowork 实战演示 ：

### 1. 文件夹收纳整理 📂

https://github.com/user-attachments/assets/dbeb0337-2d19-4b5d-a438-5220f2a87ca7

### 2. 从文件生成 PPT 📊

https://github.com/user-attachments/assets/30299ded-0260-468f-b11d-d282bb9c97f2

### 3. 从文件生成 XLSX 表格 📉

https://github.com/user-attachments/assets/f57b9106-4b2c-4747-aecd-a07f78af5dfc

---

<a id="下载与安装"></a>

## 📦 下载与安装

### 方式一：Homebrew（macOS，推荐）
Expand All @@ -91,10 +96,10 @@ brew install --cask --no-quarantine open-cowork

请访问我们的 [Release 页面](https://github.com/OpenCoworkAI/open-cowork/releases) 下载最新版本。

| 平台 | 文件类型 |
|------|----------|
| **Windows** | `.exe` |
| **macOS** (Apple Silicon) | `.dmg` |
| 平台 | 文件类型 |
| ------------------------- | -------- |
| **Windows** | `.exe` |
| **macOS** (Apple Silicon) | `.dmg` |

### 方式三：源码编译

Expand All @@ -114,11 +119,11 @@ npm run dev

Open Cowork 提供**多级沙盒保护**，确保系统安全：

| 级别 | 平台 | 技术 | 说明 |
|------|------|------|------|
| **基础** | 全平台 | 路径守卫 | 文件操作限制在工作区文件夹内 |
| **增强** | Windows | WSL2 | 命令在隔离的 Linux 虚拟机中执行 |
| **增强** | macOS | Lima | 命令在隔离的 Linux 虚拟机中执行 |
| 级别 | 平台 | 技术 | 说明 |
| -------- | ------- | -------- | ------------------------------- |
| **基础** | 全平台 | 路径守卫 | 文件操作限制在工作区文件夹内 |
| **增强** | Windows | WSL2 | 命令在隔离的 Linux 虚拟机中执行 |
| **增强** | macOS | Lima | 命令在隔离的 Linux 虚拟机中执行 |

- **Windows (WSL2)**：检测到 WSL2 后，所有 Bash 命令自动路由到 Linux 虚拟机，工作区双向同步。
- **macOS (Lima)**：安装 [Lima](https://lima-vm.io/) (`brew install lima`) 后，命令在挂载了 `/Users` 的 Ubuntu 虚拟机中运行。
Expand All @@ -129,7 +134,8 @@ Open Cowork 提供**多级沙盒保护**，确保系统安全：
- **Windows**：如已安装 WSL2，会自动检测。[安装 WSL2](https://docs.microsoft.com/zh-cn/windows/wsl/install)

- **macOS**：
如已安装 lima，会自动检测。lima安装指令如下：
如已安装 lima，会自动检测。lima安装指令如下：

```bash
brew install lima
# Open Cowork 会自动创建和管理 'claude-sandbox' 虚拟机
Expand All @@ -138,26 +144,30 @@ brew install lima
---

<a id="快速开始"></a>

## 🚀 快速开始

### 1. 获取 API Key

你需要一个 API Key 来驱动 Agent。我们支持 **OpenRouter**、**Anthropic** 以及多家高性价比的**国产大模型**。

| 服务商 | 获取 Key / Coding Plan | Base URL (必填) | 推荐模型 Model |
|-------|------------------------|-----------------|---------------|
| **OpenRouter** | [OpenRouter](https://openrouter.ai/) | `https://openrouter.ai/api` | `claude-4-5-sonnet` |
| **Anthropic** | [Anthropic Console](https://console.anthropic.com/) | 默认 | `claude-4-5-sonnet` |
| **智谱 AI** | [GLM Coding Plan](https://bigmodel.cn/glm-coding) (⚡️国产特惠) | `https://open.bigmodel.cn/api/anthropic` | `glm-4.7`, `glm-4.6` |
| **MiniMax** | [MiniMax Coding Plan](https://platform.minimaxi.com/subscribe/coding-plan) | `https://api.minimaxi.com/anthropic` | `minimax-m2` |
| **Kimi** | [Kimi Coding Plan](https://www.kimi.com/membership/pricing) | `https://api.kimi.com/coding/` | `kimi-k2` |
| 服务商 | 获取 Key / Coding Plan | Base URL (必填) | 推荐模型 Model |
| -------------- | -------------------------------------------------------------- | ---------------------------------------- | -------------------- |
| **OpenRouter** | [OpenRouter](https://openrouter.ai/) | `https://openrouter.ai/api` | `claude-4-5-sonnet` |
| **Anthropic** | [Anthropic Console](https://console.anthropic.com/) | 默认 | `claude-4-5-sonnet` |
| **智谱 AI** | [GLM Coding Plan](https://bigmodel.cn/glm-coding) (⚡️国产特惠) | `https://open.bigmodel.cn/api/anthropic` | `glm-4.7`, `glm-4.6` |
| **MiniMax** | [MiniMax 开放平台](https://platform.minimaxi.com) | `https://api.minimax.io/v1` | `MiniMax-M2.7` |
| **Kimi** | [Kimi Coding Plan](https://www.kimi.com/membership/pricing) | `https://api.kimi.com/coding/` | `kimi-k2` |

### 2. 配置应用

1. 打开应用，点击左下角 ⚙️ **设置**。
2. 填入你的 **API Key**。
3. **关键步骤**：根据上表修改 **Base URL** (如使用智谱/MiniMax等)。
4. 输入想要使用的 **Model** 名称。

### 3. 开始协作

1. **选择工作区**：选择一个文件夹，授权 Claude 在其中工作。
2. **输入指令**：
> "读取当前文件夹下的 financial_report.csv，并帮我生成一份包含 5 页幻灯片的 PPT 总结报告。"
Expand All @@ -170,12 +180,14 @@ brew install lima
```
或者前往 **系统设置 > 隐私与安全性** 点击”仍要打开”。
2. **网络连接**：对于 `WebSearch` 等联网工具，可能需要开启代理软件的“虚拟网卡 (TUN模式)”功能才能正常访问。
3. **Notion连接器使用**: 除了设置Notion token之外，还需要在根页面添加连接。更多指引请看https://www.notion.com/help/add-and-manage-connections-with-the-api。
3. **Notion连接器使用**: 除了设置Notion token之外，还需要在根页面添加连接。更多指引请看https://www.notion.com/help/add-and-manage-connections-with-the-api。

<a id="技能库"></a>

## 🧰 技能库

Open Cowork 内置技能位于 `.claude/skills/`，并支持用户自行添加/自定义技能，包含：

- `pptx`：PowerPoint 生成
- `docx`：Word 文档处理
- `pdf`：PDF 处理与表单
Expand Down
20 changes: 10 additions & 10 deletions docs/common-provider-setups.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,16 +2,16 @@

Open Cowork recommends provider setups based on the runtime/provider compatibility behavior used by the app. Some services work best with a specific compatibility protocol or with a dedicated provider tab.

| Service | Recommended setup | Base URL | Example model | Notes |
| --- | --- | --- | --- | --- |
| OpenRouter | Use the dedicated OpenRouter provider tab | `https://openrouter.ai/api/v1` | `anthropic/claude-sonnet-4-6` | Prefer the dedicated tab unless you are debugging a custom relay. |
| DeepSeek API | `Custom` + `OpenAI` | `https://api.deepseek.com/v1` | `deepseek-chat` | Use the exact model ID from DeepSeek docs. |
| Kimi Coding | `Custom` + `Anthropic` | `https://api.kimi.com/coding` | `kimi-k2-thinking` | Use the coding endpoint and verify the exact coding model ID from Moonshot docs. |
| GLM / BigModel (Anthropic route) | `Custom` + `Anthropic` | `https://open.bigmodel.cn/api/anthropic` | `glm-5` | Use this route when the endpoint path includes `/api/anthropic`. |
| Ollama | Use the dedicated Ollama provider tab | `http://localhost:11434/v1` | `qwen3.5:0.8b` | The dedicated tab supports local model refresh and discovery. |
| Gemini custom endpoint | `Custom` + `Gemini` | `https://generativelanguage.googleapis.com` | `gemini-2.5-flash` | Use the exact model ID exposed by the endpoint. |
| MiniMax | `Custom` + `OpenAI` | `https://api.minimax.chat/v1` | `MiniMax-M2.5` | Unless your gateway docs say otherwise, use OpenAI-compatible mode. |
| Generic OpenAI-compatible | `Custom` + `OpenAI` | `https://your-provider.example/v1` | `deepseek-chat` | For gateways or relays that mimic the OpenAI API. |
| Service | Recommended setup | Base URL | Example model | Notes |
| -------------------------------- | ----------------------------------------- | ------------------------------------------- | ----------------------------- | -------------------------------------------------------------------------------- |
| OpenRouter | Use the dedicated OpenRouter provider tab | `https://openrouter.ai/api/v1` | `anthropic/claude-sonnet-4-6` | Prefer the dedicated tab unless you are debugging a custom relay. |
| DeepSeek API | `Custom` + `OpenAI` | `https://api.deepseek.com/v1` | `deepseek-chat` | Use the exact model ID from DeepSeek docs. |
| Kimi Coding | `Custom` + `Anthropic` | `https://api.kimi.com/coding` | `kimi-k2-thinking` | Use the coding endpoint and verify the exact coding model ID from Moonshot docs. |
| GLM / BigModel (Anthropic route) | `Custom` + `Anthropic` | `https://open.bigmodel.cn/api/anthropic` | `glm-5` | Use this route when the endpoint path includes `/api/anthropic`. |
| Ollama | Use the dedicated Ollama provider tab | `http://localhost:11434/v1` | `qwen3.5:0.8b` | The dedicated tab supports local model refresh and discovery. |
| Gemini custom endpoint | `Custom` + `Gemini` | `https://generativelanguage.googleapis.com` | `gemini-2.5-flash` | Use the exact model ID exposed by the endpoint. |
| MiniMax | `Custom` + `OpenAI` | `https://api.minimax.io/v1` | `MiniMax-M2.7` | Unless your gateway docs say otherwise, use OpenAI-compatible mode. |
| Generic OpenAI-compatible | `Custom` + `OpenAI` | `https://your-provider.example/v1` | `deepseek-chat` | For gateways or relays that mimic the OpenAI API. |

## Common probe errors

Expand Down
51 changes: 32 additions & 19 deletions readme.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,7 @@ It provides a sandboxed workspace where AI can manage files, generate profession
---

<a id="features"></a>

## ✨ Key Features

| | MCP & Skills | Remote Control | GUI Operation |
Expand All @@ -60,30 +61,34 @@ It provides a sandboxed workspace where AI can manage files, generate profession

<a id="demo"></a>



## 🎬 Demo

See Open Cowork in action:

### 1. Folder Organization & Cleanup 📂

https://github.com/user-attachments/assets/dbeb0337-2d19-4b5d-a438-5220f2a87ca7

### 2. Generate PPT from Files 📊

https://github.com/user-attachments/assets/30299ded-0260-468f-b11d-d282bb9c97f2

### 3. Generate XLSX Spreadsheets 📉

https://github.com/user-attachments/assets/f57b9106-4b2c-4747-aecd-a07f78af5dfc

### 4. GUI Operation🖥

https://github.com/user-attachments/assets/75542c76-210f-414d-8182-1da988c148f2

### 5. Remote control with Feishu(Lark) 🤖

https://github.com/user-attachments/assets/05a703de-c0f5-407b-9a43-18b6a172fd74

---

<a id="installation"></a>

## 📦 Installation

### Option 1: Homebrew (macOS, Recommended)
Expand All @@ -99,10 +104,10 @@ brew install --cask --no-quarantine open-cowork

Get the latest version from our [Releases Page](https://github.com/OpenCoworkAI/open-cowork/releases).

| Platform | File Type |
|----------|-----------|
| **Windows** | `.exe` |
| **macOS** (Apple Silicon) | `.dmg` |
| Platform | File Type |
| ------------------------- | --------- |
| **Windows** | `.exe` |
| **macOS** (Apple Silicon) | `.dmg` |

### Option 3: Build from Source

Expand All @@ -122,11 +127,11 @@ To build the installer locally: `npm run build`

Open Cowork provides **multi-level sandbox protection** to keep your system safe:

| Level | Platform | Technology | Description |
|-------|----------|------------|-------------|
| **Basic** | All | Path Guard | File operations restricted to workspace folder |
| **Enhanced** | Windows | WSL2 | Commands execute in isolated Linux VM |
| **Enhanced** | macOS | Lima | Commands execute in isolated Linux VM |
| Level | Platform | Technology | Description |
| ------------ | -------- | ---------- | ---------------------------------------------- |
| **Basic** | All | Path Guard | File operations restricted to workspace folder |
| **Enhanced** | Windows | WSL2 | Commands execute in isolated Linux VM |
| **Enhanced** | macOS | Lima | Commands execute in isolated Linux VM |

- **Windows (WSL2)**: When WSL2 is detected, all Bash commands are automatically routed to a Linux VM. The workspace is synced bidirectionally.
- **macOS (Lima)**: When [Lima](https://lima-vm.io/) is installed (`brew install lima`), commands run in an Ubuntu VM with `/Users` mounted.
Expand All @@ -137,7 +142,8 @@ Open Cowork provides **multi-level sandbox protection** to keep your system safe
- **Windows**: WSL2 is auto-detected if installed. [Install WSL2](https://docs.microsoft.com/en-us/windows/wsl/install)

- **macOS**:
Lima is auto-detected if installed. Install command:
Lima is auto-detected if installed. Install command:

```bash
brew install lima
# Open Cowork will automatically create and manage a 'claude-sandbox' VM
Expand All @@ -146,26 +152,30 @@ brew install lima
---

<a id="quick-start"></a>

## 🚀 Quick Start Guide

### 1. Get an API Key

You need an API key to power the agent. We support **OpenRouter**, **Anthropic**, and various cost-effective **Chinese Models**.

| Provider | Get Key / Coding Plan | Base URL (Required) | Recommended Model |
|----------|-----------------------|---------------------|-------------------|
| **OpenRouter** | [OpenRouter](https://openrouter.ai/) | `https://openrouter.ai/api` | `claude-4-5-sonnet` |
| **Anthropic** | [Anthropic Console](https://console.anthropic.com/) | (Default) | `claude-4-5-sonnet` |
| Provider | Get Key / Coding Plan | Base URL (Required) | Recommended Model |
| ------------------ | ------------------------------------------------------------------ | ---------------------------------------- | -------------------- |
| **OpenRouter** | [OpenRouter](https://openrouter.ai/) | `https://openrouter.ai/api` | `claude-4-5-sonnet` |
| **Anthropic** | [Anthropic Console](https://console.anthropic.com/) | (Default) | `claude-4-5-sonnet` |
| **Zhipu AI (GLM)** | [GLM Coding Plan](https://bigmodel.cn/glm-coding) (⚡️Chinese Deal) | `https://open.bigmodel.cn/api/anthropic` | `glm-4.7`, `glm-4.6` |
| **MiniMax** | [MiniMax Coding Plan](https://platform.minimaxi.com/subscribe/coding-plan) | `https://api.minimaxi.com/anthropic` | `minimax-m2` |
| **Kimi** | [Kimi Coding Plan](https://www.kimi.com/membership/pricing) | `https://api.kimi.com/coding/` | `kimi-k2` |
| **MiniMax** | [MiniMax Platform](https://platform.minimaxi.com) | `https://api.minimax.io/v1` | `MiniMax-M2.7` |
| **Kimi** | [Kimi Coding Plan](https://www.kimi.com/membership/pricing) | `https://api.kimi.com/coding/` | `kimi-k2` |

### 2. Configure

1. Open the app and click the ⚙️ **Settings** icon in the bottom left.
2. Paste your **API Key**.
3. **Crucial**: Set the **Base URL** according to the table above (especially for Zhipu/MiniMax, etc.).
4. Enter the **Model** name you want to use.

### 3. Start Coworking

1. **Select a Workspace**: Choose a folder where Claude is allowed to work.
2. **Enter a Prompt**:
> "Read the financial_report.csv in this folder and create a PowerPoint summary with 5 slides."
Expand All @@ -177,13 +187,16 @@ You need an API key to power the agent. We support **OpenRouter**, **Anthropic**
brew tap OpenCoworkAI/tap && brew install --cask --no-quarantine open-cowork
```
2. **Network Access**: For tools like `WebSearch`, you may need to enable "Virtual Network Interface" (TUN Mode) in your proxy settings to ensure connectivity.
3. **Notion Connector**: Besides setting the integration token, you also need to add connections in a root page. See https://www.notion.com/help/add-and-manage-connections-with-the-api for more details.
3. **Notion Connector**: Besides setting the integration token, you also need to add connections in a root page. See https://www.notion.com/help/add-and-manage-connections-with-the-api for more details.

---

<a id="skills"></a>

## 🧰 Skills Library

Open Cowork ships with built-in skills under `.claude/skills/`, and supports user-added or custom skills, including:

- `pptx` for PowerPoint generation
- `docx` for Word document processing
- `pdf` for PDF handling and forms
Expand Down
28 changes: 24 additions & 4 deletions src/shared/api-model-presets.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,10 @@
export type SharedProviderType = 'openrouter' | 'anthropic' | 'custom' | 'openai' | 'gemini' | 'ollama';
export type SharedProviderType =
| 'openrouter'
| 'anthropic'
| 'custom'
| 'openai'
| 'gemini'
| 'ollama';

export type SharedCustomProtocolType = 'anthropic' | 'openai' | 'gemini';

Expand Down Expand Up @@ -102,7 +108,8 @@ export const API_PROVIDER_PRESETS: SharedProviderPresets = {
{ id: 'deepseek-reasoner', name: 'deepseek-reasoner' },
{ id: 'kimi-k2-thinking', name: 'kimi-k2-thinking' },
{ id: 'glm-5', name: 'glm-5' },
{ id: 'MiniMax-M2.5', name: 'MiniMax-M2.5' },
{ id: 'MiniMax-M2.7', name: 'MiniMax-M2.7' },
{ id: 'MiniMax-M2.7-highspeed', name: 'MiniMax-M2.7-highspeed' },
{ id: 'qwen-max', name: 'qwen-max' },
{ id: 'grok-code-fast-1', name: 'grok-code-fast-1' },
{ id: 'mistral-large-latest', name: 'mistral-large-latest' },
Expand All @@ -128,15 +135,28 @@ export const PI_AI_CURATED_PRESETS: Record<string, { piProvider: string; pick: s
},
anthropic: {
piProvider: 'anthropic',
pick: ['claude-opus-4-6', 'claude-sonnet-4-6', 'claude-haiku-4-5', 'claude-sonnet-4-5', 'claude-3-7-sonnet-latest'],
pick: [
'claude-opus-4-6',
'claude-sonnet-4-6',
'claude-haiku-4-5',
'claude-sonnet-4-5',
'claude-3-7-sonnet-latest',
],
},
openai: {
piProvider: 'openai',
pick: ['gpt-5.4', 'gpt-5.4-pro', 'gpt-5-mini', 'gpt-5-nano', 'gpt-5.3-codex', 'o3', 'gpt-4.1'],
},
gemini: {
piProvider: 'google',
pick: ['gemini-3.1-pro-preview', 'gemini-3-flash-preview', 'gemini-3.1-flash-lite-preview', 'gemini-2.5-pro', 'gemini-2.5-flash', 'gemini-2.5-flash-lite'],
pick: [
'gemini-3.1-pro-preview',
'gemini-3-flash-preview',
'gemini-3.1-flash-lite-preview',
'gemini-2.5-pro',
'gemini-2.5-flash',
'gemini-2.5-flash-lite',
],
},
};

Expand Down
Loading
Loading