dsh-better-reasoning-effort
Run the following command in DeepSeek Harness:
dsh plugin install HaoyueQin/dsh-better-reasoning-effort
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, install this plugin by running the command dsh plugin install HaoyueQin/dsh-better-reasoning-effort, with the full source address at https://github.com/HaoyueQin/dsh-better-reasoning-effort.
About this plugin
The official model editor in DeepSeek Harness deliberately restricts configuration for third-party models' reasoning effort and input modalities, preventing users from enabling thinking levels or image input and limiting advanced functionality. This plugin solves the problem by injecting a custom editor block directly into the model page, allowing users to customize these settings. Its core capabilities include auto-adaptation of recommended configurations from a model knowledge base and protocol inference, zoned suggestion display, and one-click application. It is designed for all developers using DeepSeek Harness for model inference, especially researchers and engineers relying on third-party APIs, enabling them to easily optimize configurations and unlock the full potential of their models.
Screenshots
Use Cases
- Enable thinking level and image input support for third-party models
- Auto-fill recommended inference settings from a model knowledge base
- Edit configurations directly on the official model page without leaving the interface
Best For
- Developers using DeepSeek Harness for model inference
- Researchers and engineers configuring third-party API models
- Users needing to optimize model configurations for better inference efficiency
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.