AI Agent Hub
Back to plugins
🤖

dsh-llm-cost

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install chenyinrusi/dsh-llm-cost

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

To install this plugin in DeepSeek Harness, please use the command: dsh plugin install chenyinrusi/dsh-llm-cost. Source: https://github.com/chenyinrusi/dsh-llm-cost

About this plugin

DeepSeek Harness (DSH) is a powerful model inference tool, but it has long lacked a critical "cost" dimension—how to translate token usage into dollar expenditures. The dsh-llm-cost plugin is designed to solve this problem by seamlessly integrating into DSH's session view, converting usage in real-time into monetary value to make costs transparent. Its core capability lies in its ability to automatically maintain pricing tables, leveraging web search and LLM extraction to ensure the timeliness of pricing data, thereby preventing budget control issues caused by price fluctuations. This is an indispensable utility tool for developers, researchers, or anyone running large-scale LLM services who need precise control over API expenses. It solves the "unknown spending" pain point and provides clear visual feedback to assist in decision-making.

Use Cases

  • Scenarios requiring precise control over API expenditure per conversational turn.
  • Scenarios needing automatic maintenance of the latest LLM API pricing tables.
  • Scenarios requiring aggregation and analysis of inference costs by model.

Best For

  • Developers running DeepSeek Harness for model inference.
  • Researchers aiming to optimize API call budgets.
  • Project managers needing detailed token usage statistics and cost breakdowns.