Ebony-Vinyl/dsh-our-free-model
在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费,不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, DeepSeek V4.1 Flash and Kimi K3 among them. Completely free, with no usage cap.
What it solves
This plugin removes the friction of accessing advanced AI models within the DeepSeek Harness (dsh) environment. It eliminates the need for user accounts, API keys, or manual configuration, providing immediate, free, and uncapped access to frontier models like DeepSeek V4.1 Flash and Kimi K3.
How it works
Once installed, the plugin automatically connects to public gateways (such as OpenCode and Kilo AI) to fetch available models. It handles all authentication and routing in the background. It also provides a local OpenAI-compatible server, allowing other tools on your machine to use these models via a local port. The plugin includes built-in features to compress tool outputs (saving tokens), manage "thinking" effort for reasoning models, and organize models into fallback chains for higher reliability.
Who it’s for
It is designed for users of the DeepSeek Harness (dsh) platform who want to experiment with or use high-end AI models without the overhead of managing subscriptions, keys, or complex setup processes. It is particularly useful for developers who want to integrate these models into local workflows or CLI tools.
Highlights
- Zero-config access: No login, sign-up, or API keys required.
- Transparent sourcing: Connects directly to public gateways without third-party middleman proxies.
- Token efficiency: Automatically compresses tool outputs (like git diffs) to save tokens without sacrificing model performance.
- Local Gateway: Provides an OpenAI-compatible local port to bridge these models to other applications.
- Built-in Management: Includes an in-app update system, usage dashboard, and real-time announcement feed.
- Reasoning Control: Offers granular control over "thinking" effort for supported models to balance output quality and budget.
Related
- Project
- Project
- Project
- Project
- Project