AI Agent Hub
Back to plugins
🤖

dsh-llm-newapi

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install wenzetan/dsh-llm-newapi

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

Install this plugin in DeepSeek Harness by running 'dsh plugin install wenzetan/dsh-llm-newapi' and refer to the source code repository at https://github.com/wenzetan/dsh-llm-newapi.

About this plugin

Integrating diverse LLM providers into development tools often poses compatibility hurdles, particularly when connecting gateways like NewAPI to DeepSeek Harness. The dsh-llm-newapi plugin tackles this by seamlessly adding NewAPI support to dsh without modifying its core, enabling developers to effortlessly expand model inference capabilities. Its core strengths lie in implementing the LlmAdapter seam for model discovery, offering a settings page to configure API keys and gateway addresses, and featuring a parameter-matching engine that auto-identifies and filters models, along with reasoning effort presets for streamlined management. This plugin is ideal for developers, researchers, or anyone needing to integrate NewAPI or OpenAI-compatible gateways into DeepSeek Harness, providing a efficient and user-friendly solution for flexible LLM service usage.

Use Cases

  • Configuring and using NewAPI-compatible LLM services within DeepSeek Harness.
  • Automatically discovering and filtering model lists for chat completions.
  • Managing API keys and gateway addresses via a settings page without core modifications.

Best For

  • AI developers needing to extend LLM provider options in DeepSeek Harness.
  • Researchers aiming to integrate various OpenAI-compatible gateways for experiments.
  • Technical teams responsible for deploying and managing LLM inference services.