AI Agent Hub
Back to plugins
🤖

dsh-models-dev-reasoning

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install aerince/dsh-models-dev-reasoning

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

To install this plugin in DeepSeek Harness, run dsh plugin install aerince/dsh-models-dev-reasoning, with the full source address at https://github.com/aerince/dsh-models-dev-reasoning.

About this plugin

When using DeepSeek Harness, users often add third-party or custom models that frequently lack reasoning efforts configuration, preventing the llm-pi-ai model selector from displaying thinking tiers and limiting inference completeness. The dsh-models-dev-reasoning plugin solves this by automatically fetching model data from models.dev and intelligently filling in the missing reasoning efforts for models in DeepSeek Harness. It operates non-intrusively, preserving existing configurations, and includes fallback and caching mechanisms for enhanced reliability. This plugin is ideal for developers or advanced users who manage custom models via hand-fill, proxies, or NewAPI scenarios and seek to boost their models' inference capabilities without disruption.

Use Cases

  • Adding thinking tiers for hand-filled models.
  • Automatically completing inference configurations in proxy model scenarios.
  • Ensuring complete model inference functionality when integrating with NewAPI.

Best For

  • Developers using custom models.
  • Advanced users managing third-party models.
  • Administrators seeking to optimize model inference configurations.