dsh-nous-portal-free-provider
Run the following command in DeepSeek Harness:
dsh plugin install jiesou/dsh-nous-portal-free-provider
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install jiesou/dsh-nous-portal-free-provider in the DeepSeek Harness terminal to install this plugin; full source at https://github.com/jiesou/dsh-nous-portal-free-provider
About this plugin
Nous Portal (Hermes Agent) offers a free OAuth tier with models like Hy3 and Ox Alpha at zero cost, yet those models were invisible inside dsh. dsh-nous-portal-free-provider bridges the Portal free endpoint into dsh so you can mix free and paid inference in a single workflow without juggling two separate tools.
The plugin discovers the current free-model set from the public /v1/models endpoint at startup, filters to prompt and completion price of zero, and follows upstream metadata (context window, multimodal input, reasoning efforts). Authentication uses a standalone device-code CLI; credentials live in the dsh credential service, and the entire refresh-token rotation (read, POST, write-back) runs inside a modifyRecord lock to prevent concurrent reuse detection. Reasoning effort exposes only the levels the upstream actually accepts, with defaults and mandatory flags delegated entirely to the provider.
Built for independent developers, researchers, and small teams who prototype, prompt, and reason daily inside dsh and want the free tier to shave down experimentation costs. As long as a browser can log in to Portal, everything else stays in the terminal.
Use Cases
- Call Hy3, Ox Alpha and other free-tier models inside dsh for daily prototyping
- Use the free tier to cut inference costs during iteration and reserve paid models for production
- Automatically pick up new free models when Portal rotates the set, no manual list edits needed
Best For
- Independent developers who prompt and reason daily inside dsh
- Researchers and academic teams seeking low-cost experimentation
- Small teams that want to mix free and paid inference in a single workflow
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.