deepseek-peak-valley-router
Run the following command in DeepSeek Harness:
dsh plugin install Eligahyu/deepseek-peak-valley-router
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, please run the command `dsh plugin install Eligahyu/deepseek-peak-valley-router` or directly enter the full source code URL https://github.com/Eligahyu/deepseek-peak-valley-router.
About this plugin
DeepSeek V4 API has introduced a peak and valley pricing strategy where peak hours are twice as expensive as off-peak hours, creating a challenge for cost optimization. To balance service quality with budget constraints, this plugin acts as an intelligent traffic controller. It automatically detects whether the current time—based on your configured reference time zone (defaulting to Beijing time)—is a high-cost peak or a low-cost off-peak period. During peak hours, the plugin seamlessly routes requests to the cost-effective deepseek-v4-flash model; conversely, during off-peak hours, it switches to the high-performance deepseek-v4-pro model, ensuring you get the best output quality for a lower price. Crucially, it prioritizes your explicit choices, ensuring you never lose control over your model selection unless you intentionally change it. This makes it an essential tool for developers and users looking to reduce API expenses without sacrificing performance.
Use Cases
- Performing rapid code generation during peak hours to save costs
- Automating batch tasks to reduce API fees
- Development workflows requiring intelligent model scheduling
Best For
- Developers seeking to save DeepSeek API costs
- AI application engineers needing automatic model switching
- Enterprise users using DeepSeek V4 who prioritize budget
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.