oebeliever/dsh-image2-draw ↗★ 0
dsh-image2-draw
Image2 (gpt-image-2) generation plugin for DeepSeek Harness with simple relay settings, in-conversation previews, a chat-dock studio (text-to-image / image-to-image / multi-view character consistency) and a reusable prompt/flow preset library. Community fork of JuneLearn/dsh-image2-draw with gateway compatibility fixes.
安装
npx -p @deepseek-ai/dsh dsh plugin --profile web add github:oebeliever/dsh-image2-draw说明文档
阅读完整 README ↗Usage
- Settings → Plugins → Plugin config → Image2 生图.
- Fill in
API KeyandbaseURL(e.g.https://example.com/v1) from your relay. - Adjust model / edit endpoint / timeout only if needed; defaults usually work.
- Save, start a new session, and ask the model to call
image2-generate, or pass reference-image paths toimage2-edit. - Prefer the graphical way for image inputs: open the 🎨 Image2 生图工作台 above the input box, pick a tab (文生图 / 图生图 / 多视角人物), drop images in and generate.
Example prompts:
Call image2-generate to create a portrait cinematic poster of a future city at high quality.
Call image2-edit with D:\images\room.png and restyle the room with light Japanese wood while preserving the layout.
Whether the relay actually supports gpt-image-2, editing, custom sizes and quality
levels is up to the provider. HTTP 400 / 404 → check the provider's model name and
Images endpoint format. Size errors → most gateways accept only auto / 1024x1024 /
1536x1024 / 1024x1536 / 3840x2160; pick a preset instead of a custom WIDTHxHEIGHT.
region-unavailable → the relay blocks your exit IP region;
switch to an overseas proxy or another provider.