dsh-highres-vision
For the DeepSeek Harness native vision model deepseek-v4-flash-vision-exp, raises image admission limits to 32 MiB / 8192 px / 600 images and adds a highres_read tool that tiles large images, then returns the whole image plus 800x800 tiles through the host read_image tool.
为 DeepSeek Harness 原生视觉模型 deepseek-v4-flash-vision-exp 提供高清识图:将图片准入上限提升至 32 MiB / 8192 px / 600 张,并新增 highres_read 工具,将大图切为整图与 800x800 分块后经宿主 read_image 注入模型。
How to install
dsh plugin add github:azwosile/dsh-highres-vision About
dsh-highres-vision 大肥鱼出来卖个萌 🐳 鲸鱼娘登场喵~ 如果这个小插件帮到了你,记得给仓库点个 ⭐ Star 喵! 星星越多,鲸鱼娘越开心,也会更有动力继续维护这个插件喵~ 💙 (求求啦喵~) ⚠️ **本插件专供 deepseek-v4-flash-vision-exp 模型使用。** 该模型是 DeepSeek 的视觉多模态模型,支持图片输入; 本插件针对它的 800×800 等效缩放、单图 32MiB / 8192px 上限做了高清分块增强。 **其他模型通常不支持图片输入,或无法获得同样的高清识图效果。** DeepSeek Harness 高清识图增强插件(v0.2.0)。 (本身也是由 deepseek v4 flash vision exp 开发()) 整合两个能力: 1. **放宽图片限制** 单图 32 MiB 单边 8192px 单请求 600 张 inline 总量 64 MiB base64 累计 44 MiB 2. **高清分块识图** 注册 highres_read 工具 自动定位当前会话最近一张用户图片 生成整图 + ≤800×800 高清分块(overlap 自动:40/80/120px) 通过宿主 read_image 把整图和每个分块注入模型 **不覆盖宿主 read_image**,read_image 保持原样 age…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 8
- mo. downloads
- –
- Last push
- 2026-08-25
- Created
- 2026-08-24
Links
Basic safety check
- Findings
- curated 收录但无 npm 包/安装命令
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- deepseek-harness, dsh-plugin, image-recognition, vision
Related plugins
dsh-vision-router
ysr666/dsh-vision-router
Free vision for text-only agents: built-in keyless vision chain plus pixel tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots); paste an image to use it.
modlens
liustack/modlens
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
dsh-web
zhu1090093659/dsh-web-ui/tree/main/packages/dsh-tool-describe-image
A `describe_image` vision tool for text-only models: images (local path, URL, attachment) go to a configurable OpenAI-compatible vision endpoint and only the returned text enters the session.
dsh-vision-toolkit
Anionex/dsh-vision-toolkit
Vision for text-only models: paste an image and the model switches to a Vision Toolkit variant for image Q&A, multi-image comparison, long-screenshot OCR, screenshot-to-UI reproduction, element grounding, and pixel diff. No API key by default — images are processed by the author-hosted free service, 100 per machine per day; configurable to your own provider.