AI Agent Hub
Back to plugins
🤖

dsh-providers

Model Inference Updated 2026.08.22

Run the following command in DeepSeek Harness:

dsh plugin install tyql688/dsh-providers

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

Run dsh plugin install tyql688/dsh-providers in the DeepSeek Harness terminal to install (source: https://github.com/tyql688/dsh-providers).

About this plugin

Juggling multiple model providers in DeepSeek Harness—Anthropic, OpenRouter, DeepSeek, GLM and more—means login sessions, model lists, spend, and plan quotas are scattered across different places. dsh-providers consolidates them: manage OAuth and API-key accounts side by side, refresh model catalogs from curated sources or OpenAI-compatible endpoints, and track daily tokens, costs, and reset countdowns at a glance.\n\nIt also breaks down per-session stats—tokens, cost, cache hit rate, first-token latency, subagent calls, and tool-invocation error rates—and lets you hand-enter prices for any model the catalog does not cover, re-pricing figures instantly. Nothing polls in the background; data is fetched only when you open a panel, credentials stay local, and OAuth tokens never reach the browser. Built for developers who run three or more LLM providers and need tight control over spend and quota.

Use Cases

  • Manage logins and model lists across multiple LLM providers from one place
  • Track daily token spend, cost, and cache hit rate with a 14-day trend
  • Monitor subscription quota windows to catch rate-limit resets before they hit

Best For

  • Developers running three or more LLM providers concurrently
  • Teams needing fine-grained cost and quota control over API usage
  • Engineers doing multi-model inference with pi-ai in DeepSeek Harness