AI Agent Hub
Back to plugins
dsh-concurrency-meter preview

dsh-concurrency-meter

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install Wanbinyu/dsh-concurrency-meter

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

In DeepSeek Harness, you can install this plugin using the command dsh plugin install Wanbinyu/dsh-concurrency-meter, with the full source code available at https://github.com/Wanbinyu/dsh-concurrency-meter.

About this plugin

Real-time monitoring of request concurrency is crucial for optimizing performance and troubleshooting in model inference services, but intrusive tools may introduce additional risks. The dsh-concurrency-meter plugin provides a safe, read-only monitoring solution for DeepSeek Harness Web, seamlessly integrating into the call chain to track request states with minimal overhead.

The plugin's core capability lies in comprehensive and non-disruptive observation: it displays current active requests, peak concurrency, and supplier-specific statistics (such as success, failure, abort, etc.) in real time, while supporting configurable concurrency threshold alerts and statistics reset. Importantly, it does not intercept, queue, or modify any requests, ensuring the monitoring process remains entirely transparent.

Whether you are a developer debugging API call flows, an operations professional monitoring service health, or a researcher analyzing model load, dsh-concurrency-meter offers intuitive insights. It is particularly suitable for users who wish to safely monitor concurrency behavior in production environments without introducing any potential interference.

Screenshots

Use Cases

  • Monitor model request concurrency status in real time
  • Check peak concurrency when debugging performance bottlenecks
  • Analyze request handling performance across different suppliers

Best For

  • Developers debugging API call flows
  • Operations professionals monitoring service health
  • Researchers analyzing model load behavior