English

[!IMPORTANT] This is not affiliated with, officially maintained by, or endorsed by TokenRhythm.

Integrate TokenRhythm models into GitHub Copilot Chat as a VS Code extension.

Usage

1. Set API Key: Ctrl+Shift+P → TokenRhythm: Set TokenRhythm API Key
2. Show Models: Click the settings icon in the model picker → Language Models panel → set your desired models to Visible
3. Select Model: In the Copilot Chat bottom model picker, choose a "TokenRhythm" model
4. Start chatting

Advanced Token Usage Indicator

Once installed, the status bar shows the current context usage and cumulative input/output token counts for TokenRhythm models. DeepSeek models and models that return cache metrics via the OpenAI-compatible format also display the cumulative cache hit count and cache hit rate in the tooltip.

The status bar only appears while you are actually using a TokenRhythm model: it stays hidden on startup and when other chat model providers are in use, and auto-hides after 60 seconds of inactivity.

You can control this indicator via the tokenrhythm.enableThirdPartyTokenIndicator setting (default: true). When disabled, only the native Copilot token indicator remains visible.

[!NOTE] Whether non-DeepSeek models display cache data depends on whether the model API returns cache metrics in an OpenAI-compatible format. This does not indicate whether the model supports caching — caching support depends on TokenRhythm.

Git Commit Messages

Click the magic wand button in the Source Control (SCM) panel to auto-generate a commit message.

You can configure the model, language, number of recent commits to reference, and whether to attach context files.

Model Temperature Presets

Quickly switch temperature presets via Ctrl+Shift+P → TokenRhythm: Set Model Temperature Preset.

Built-in presets:

You can also configure tokenrhythm.temperature and tokenrhythm.top_p directly in settings.json (requires tokenrhythm.modelPreset set to "custom").

Extended Vision Understanding

This extension adds extended vision understanding capability to text-only models that do not natively support vision. When you send a message with an image to these models, they can call a vision-capable model to describe the image, and then answer based on that description.

You can configure the default vision model and whether to enable thinking when describing images. By default, Kimi K2.6 is used to describe images.

Model List

The extension ships with built-in definitions for the following TokenRhythm chat models (sourced from the model page):

All models support the OpenAI-compatible protocol. The Responses API column shows which models additionally support the Responses protocol (supports_responses=true); most models also support Anthropic (supports_anthropic=true). Protocol capability is detected dynamically at startup from GET /v1/models — no model IDs are hardcoded. In auto mode, priority: Responses (if enableResponsesApi enabled) > Anthropic (if enableAnthropicApi enabled) > OpenAI. ¹ kimi-k2.7-code does not support temperature/top_p parameters. ² qwen3.7-max does not support the Anthropic protocol (supports_anthropic=false).

[!TIP] Automatic model discovery is enabled by default: the extension fetches the live model list from GET /v1/models and hides models that are not available on your account. Image-generation models (qwen-image-2.0, wan2.7-image) are excluded from the picker.

Configuration

Available in settings.json:

{
  "tokenrhythm.apiMode": "auto",
  "tokenrhythm.commitLanguage": "auto",
  "tokenrhythm.commitModel": "deepseek-v4-flash",
  "tokenrhythm.commitMessagePrompt": "",
  "tokenrhythm.requestTimeout": 600000,
  "tokenrhythm.recentCommitsCount": 10,
  "tokenrhythm.commitIncludeCommitDiff": false,
  "tokenrhythm.commitAttachContextFiles": true,
  "tokenrhythm.enableAutoModelDiscovery": true,
  "tokenrhythm.enableThirdPartyTokenIndicator": true,
  "tokenrhythm.enableResponsesApi": false,
  "tokenrhythm.enableAnthropicApi": false
}

[!NOTE] Models with switchable thinking (e.g., DeepSeek, Qwen) provide reasoning effort levels such as Disabled/High/Maximum.

Build

npm install
npm run compile
npm run build      # packages extension.vsix

License

AGPL-3.0 License. This project builds upon the architecture of opencode-go-copilot (MIT) and oai-compatible-copilot (MIT).

中文

[!IMPORTANT] 本插件与 TokenRhythm 无关，也未获得其官方维护或认可。

将 TokenRhythm 模型集成到 GitHub Copilot Chat 的 VS Code 插件。

使用

1. 设置 API Key：Ctrl+Shift+P → TokenRhythm: Set TokenRhythm API Key
2. 显示模型：在模型选择器中点击设置图标 → 语言模型 面板 → 将需要使用的模型显示
3. 选择模型：在 Copilot Chat 底部模型选择器中选择 "TokenRhythm" 下的模型
4. 开始对话

高级 Token 用量指示器

安装后，使用 TokenRhythm 提供的模型时，状态栏会显示当前上下文用量与累计输入/输出 Token 量。DeepSeek 和通过 OpenAI 格式返回缓存用量的模型还会显示累计缓存命中量与缓存命中率。

状态栏仅在您实际使用 TokenRhythm 模型时显示：启动时隐藏、使用其他模型提供商的模型时不显示，停止使用（空闲 60 秒）后自动隐藏。

可通过 tokenrhythm.enableThirdPartyTokenIndicator 设置（默认 true）控制此高级 Token 指示器。关闭后仅显示 Copilot 原生 Token 指示器。

[!NOTE] 非 DeepSeek 的模型是否显示缓存数据取决于模型接口是否通过 OpenAI 格式返回缓存数据，这并不代表此模型是否支持缓存。模型对于缓存的支持情况取决于 TokenRhythm。

Git 提交消息

在源代码管理（SCM）面板中点击魔法棒按钮，自动生成 Git 提交消息。

可在配置里配置使用的模型、语言、参考的最近提交数量以及是否附加上下文文件。

扩展视觉理解

本插件为不支持视觉理解的纯文本模型添加了扩展视觉理解功能，当你向这些模型发送带有图片的信息时，他们可以调用支持视觉理解的模型为图片输出描述，然后再回答。

通过配置文件可更改默认使用的模型以及是否在描述图片时启用思考。默认情况下，将使用 Kimi K2.6 描述图片。

模型列表

扩展内置了以下 TokenRhythm Chat 模型定义（来源：模型页）：

所有模型均支持 OpenAI 兼容协议。Responses API 列标注哪些模型额外支持 Responses 协议（supports_responses=true）；大多数模型也支持 Anthropic（supports_anthropic=true）。协议能力在启动时从 GET /v1/models 动态探测——不硬编码模型 ID。auto 模式下优先级：Responses（若开启 enableResponsesApi）> Anthropic（若开启 enableAnthropicApi）> OpenAI。 ¹ kimi-k2.7-code 不支持设置 Temperature/Top-p 参数。 ² qwen3.7-max 不支持 Anthropic 协议（supports_anthropic=false）。

[!TIP] 自动模型发现默认开启：扩展会从 GET /v1/models 拉取实时模型列表，隐藏你账号下不可用的模型。图片生成模型（qwen-image-2.0、wan2.7-image）不会出现在选择器中。

调整模型温度

通过 Ctrl+Shift+P → TokenRhythm: Set Model Temperature Preset 快速切换温度预设。

内置 4 个预设档位：

也可在 settings.json 中直接配置 tokenrhythm.temperature 和 tokenrhythm.top_p（需将 tokenrhythm.modelPreset 设为 "custom"）。

配置

可在 settings.json 中配置：

{
  "tokenrhythm.apiMode": "auto",
  "tokenrhythm.commitLanguage": "auto",
  "tokenrhythm.commitModel": "deepseek-v4-flash",
  "tokenrhythm.commitMessagePrompt": "",
  "tokenrhythm.requestTimeout": 600000,
  "tokenrhythm.recentCommitsCount": 10,
  "tokenrhythm.commitIncludeCommitDiff": false,
  "tokenrhythm.commitAttachContextFiles": true,
  "tokenrhythm.enableAutoModelDiscovery": true,
  "tokenrhythm.enableThirdPartyTokenIndicator": true,
  "tokenrhythm.enableResponsesApi": false,
  "tokenrhythm.enableAnthropicApi": false
}

[!NOTE] 支持切换思考模式的模型（如 DeepSeek、Qwen）提供禁用思考/高/极高等推理强度选项。

编译

npm install
npm run compile
npm run build      # 打包为 extension.vsix

许可

AGPL-3.0 许可。本项目基于 opencode-go-copilot（MIT）与 oai-compatible-copilot（MIT）的架构实现。