dsh-model-provider-label
Run the following command in DeepSeek Harness:
dsh plugin install haiyoucuv/dsh-model-provider-label
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install haiyoucuv/dsh-model-provider-label in your terminal to install; source code is available at https://github.com/haiyoucuv/dsh-model-provider-label
About this plugin
When you configure multiple providers in DeepSeek Harness, different providers often carry models with identical names — say, two providers both list GPT-5. The default model selector shows only the model name, leaving you with no way to tell them apart. dsh-model-provider-label solves this small but nagging UX gap: it prepends the provider display name to the model name (e.g. DeepSeek · DeepSeek V4 Flash) so every option is instantly distinguishable.
The implementation is deliberately minimal. It leverages the DSH slot system by registering the same slot at a lower priority, replacing only the render layer — dropdown behavior, reasoning-level selection, model directory loading, and error handling all remain 100% host-service driven. There is no extra state, no logic fork; you simply get a provider prefix in front of each label while every underlying pipeline stays untouched.
If you run two or more providers with overlapping model names, this plugin (or its successor dsh-chat-tweaks) is worth a look. Note that dsh-model-provider-label is now deprecated; the provider-label feature has been folded into dsh-chat-tweaks, which continues to receive updates.
Screenshots
Use Cases
- Disambiguate same-named models across multiple providers at a glance
- Identify which provider each model belongs to in the dropdown
- Enhance the default model selector readability with a minimal change
Best For
- DSH users running two or more providers simultaneously
- Developers hitting cross-provider model name collisions
- Power users who want richer context in the model picker
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.