dsh-model-center
Run the following command in DeepSeek Harness:
dsh plugin install artwar2020/dsh-model-center
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install artwar2020/dsh-model-center in your terminal (source at https://github.com/artwar2020/dsh-model-center) and restart the DSH web server to see the Model Center card.
About this plugin
When juggling multiple model routes, free-model availability, pricing, and usage metrics tend to live on separate dashboards—switching a model means tab-hopping across sites. dsh-model-center pulls all of this into a single card inside the DeepSeek Harness web UI, so you can assess and switch models without leaving the interface.
It ingests every :free model from OpenRouter, scores each one on a 0–100 scale using three signals—Artificial Analysis intelligence index, Arena Elo, and parameter count—and assigns a 1–5 star rating with auto-derived specialty tags such as reasoning, code, and multimodal. A one-click probe fires a max_tokens=1 request to verify real availability, caching results for five minutes and never touching paid endpoints. Pricing tables from TeamoRouter, live balance, customizable domestic price lists, and historical usage stats (carrying over the dsh-usage-stats backend) all appear in the same card, with context-window lengths auto-enriched from the catalog.
Built for solo developers and small teams who run multi-route inference in DSH and want to save costs with free models without guessing which ones actually work. Every API route is restricted to loopback access with a triple-check security fence, and API keys are sent only as Bearer headers to upstream services—never logged or rendered in the UI.
Use Cases
- Survey every free model's rating, specialty tags, and context window in a single DSH card
- Probe a free model's real availability with one click before switching your active route
- Check TeamoRouter pricing, account balance, and usage history without leaving the interface
Best For
- Solo developers running multi-route inference on DeepSeek Harness
- Engineering teams that cut inference costs with free models but need verified availability
- Platform maintainers who want model ratings, pricing, and usage consolidated in one dashboard
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.