AI Agent Hub
Back to plugins
🧰

dsh-llm-balance

Web Tools Updated 2026.09.09

Run the following command in DeepSeek Harness:

dsh plugin install JonyChan1350/dsh-llm-balance

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install JonyChan1350/dsh-llm-balance in your terminal to install; the source is at https://github.com/JonyChan1350/dsh-llm-balance . Restart the dsh web server afterward to activate.

About this plugin

dsh users who have multiple LLM providers configured often break their workflow to manually check each provider console for remaining credits. dsh-llm-balance brings that check right into the chat interface, rendering a compact balance strip directly below the input box that auto-refreshes every 60 seconds by default, with a manual refresh link on demand.

The strip is fully driven by your existing dsh configuration: add a new model route and it appears on the next poll; remove one and it disappears immediately. No restart, no config file edits. For providers with built-in balance endpoints such as DeepSeek, OpenRouter, SiliconFlow, Moonshot/Kimi, MiniMax, StepFun, and Zhipu/GLM, clicking a balance figure opens the corresponding recharge page in a new tab. Providers without a public balance endpoint, like Qwen, OpenAI, and Anthropic, are shown with a clear no-API label rather than being silently hidden. API keys are always resolved host-side through dsh credentials; the browser only ever sees the fetched balance values, keeping sensitive tokens out of the frontend.

Best suited for dsh users running two or more LLM routes who want at-a-glance balance visibility and one-click access to top-up pages. No extra setup is required beyond installing the plugin and restarting the dsh web server.

Use Cases

  • Monitor all provider balances in one compact strip without switching browser tabs
  • Click a balance figure to jump straight to that provider's top-up page
  • Auto-sync the strip when you add or remove a provider, no restart needed

Best For

  • dsh users running two or more LLM provider routes
  • Developers who check credits and recharge on a regular cadence
  • AI application teams looking to cut manual balance-checking overhead