dsh-rembg
Run the following command in DeepSeek Harness:
dsh plugin install mengruoa/dsh-rembg
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install mengruoa/dsh-rembg in the DeepSeek Harness terminal to install this plugin; the full source code is available at https://github.com/mengruoa/dsh-rembg .
About this plugin
In a DSH workflow, stripping image backgrounds typically means spinning up a separate service or writing standalone scripts to call back. dsh-rembg embeds this ability directly into the DSH plugin system so a model can remove a background with a single tool call, eliminating cross-process and cross-environment overhead.
The plugin offers two inference modes: CPU and local NVIDIA GPU. You switch modes in the Web GUI settings panel, click Initialize Environment to install the matching Python dependencies and pull a model, and everything is stored within the plugin install directory without touching system Python. GPU mode requires a working NVIDIA driver and nvidia-smi; CPU mode has no extra hardware prerequisite. Once initialized, you simply pass the absolute path of an image and optionally a model name to receive the output file.
It is well suited for developers and designers who need to clean up image backgrounds locally and offline, as well as for users who own an NVIDIA card and want GPU-accelerated inference. The entire pipeline runs on-premises with no cloud API dependency, keeping your data on your own machine.
Use Cases
- Clean up product photos or asset backgrounds locally at scale
- Perform offline background removal without any cloud API dependency
- Accelerate bulk image background removal using an NVIDIA GPU
Best For
- Designers and developers who need local offline image background cleanup
- Users with an NVIDIA GPU seeking accelerated inference
- Automation engineers integrating image processing into DSH workflows
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.