示例用 -imagor-unsafe 关闭安全校验,生产环境需自行配置签名与安全策略;视频缩略图需另用 imagorvideo(依赖 ffmpeg C 绑定);默认监听 8000 端口,需与 -p 映射一致。

imagor
cshumApache-2.0活跃维护基于 libvips 的高性能图像处理服务器和 Go 库,支持裁剪、缩放、格式转换,可 Docker 部署,适合给应用提供安全快速的图片处理接口。
Fast, secure image processing server and Go library, using libvips
docker run -p 8000:8000 shumc/imagor -imagor-unsafe -imagor-auto-webpREADME 未说明具体显存或 GPU 要求;以 Docker 镜像方式运行,依赖 libvips。
基于 libvips 的高性能图像处理服务器与 Go 库,兼容 thumbor URL 语法
基于 libvips 的高性能图像处理服务器和 Go 库,支持裁剪、缩放、格式转换,可 Docker 部署,适合给应用提供安全快速的图片处理接口。
核心能力
- 用 libvips 与 vipsgen 绑定做图像处理
- 实现 libvips 流式处理以支持并行管线
- 兼容 thumbor URL 语法,可直接替换
- 提供 HTTP 服务与 Docker 镜像
- 支持滤镜链、水印、格式与质量参数
容易踩的坑
- v1.10.1最新正式版2026-11-07 · 2026-11-07 13:04 UTC
What's Changed
- feat: add IMAGOR_AUTO_JXL config option by @cshum in https://github.com/cshum/imagor/pull/881
- feat: keep the gain map when metadata is stripped, decode it smaller by @cshum in https://github.com/cshum/imagor/pull/875
- fix: keep the gain map through orient and strip_exif by @cshum in https://github.com/cshum/imagor/pull/873
- fix: save and transform the gain map of iPhones by @cshum in https://github.com/cshum/imagor/pull/874
- perf: decode the gain map as small as the resize needs by @cshum in https://github.com/cshum/imagor/pull/876
Full Changelog: https://github.com/cshum/imagor/compare/v1.10.0...v1.10.1
- v1.10.02026-11-06 · 2026-11-06 03:15 UTC
查看更新说明
What's Changed
- feat: UltraHDR gain map support with vipsgen 2.1.2 by @cshum in https://github.com/cshum/imagor/pull/865
- perf(svg): reduce sanitization CPU cost and allocations by @pixel365 in https://github.com/cshum/imagor/pull/866
- fix(vips): keep the last column and row of a path crop by @cshum in https://github.com/cshum/imagor/pull/869
- feat(storage): skip Storage by image prefixes by @cshum in https://github.com/cshum/imagor/pull/868
New Contributors
- @pixel365 made their first contribution in https://github.com/cshum/imagor/pull/866
Full Changelog: https://github.com/cshum/imagor/compare/v1.9.14...v1.10.0
- v1.9.142026-10-24 · 2026-10-24 18:36 UTC
查看更新说明
What's Changed
- Update Dockerfile to golang v1.27.3 by @EyePulp in https://github.com/cshum/imagor/pull/862
- build: bump vipsgen to v2.1.0
Full Changelog: https://github.com/cshum/imagor/compare/v1.9.13...v1.9.14
- v1.9.132026-10-20 · 2026-10-20 15:38 UTC
查看更新说明
What's Changed
- fix(filestorage): confine storage paths to the base directory by @choksi2212 in https://github.com/cshum/imagor/pull/858
- fix(vipsprocessor): fail a request whose source read failed mid-render by @cshum in https://github.com/cshum/imagor/pull/860
New Contributors
- @choksi2212 made their first contribution in https://github.com/cshum/imagor/pull/858
Full Changelog: https://github.com/cshum/imagor/compare/v1.9.12...v1.9.13
- v1.9.122026-10-17 · 2026-10-17 12:48 UTC
查看更新说明
What's Changed
- fix(s3storage): buffer the body when the blob size is unknown by @fclairamb in https://github.com/cshum/imagor/pull/855
- refactor: migrate to vipsgen v2 by @cshum in https://github.com/cshum/imagor/pull/856
New Contributors
- @fclairamb made their first contribution in https://github.com/cshum/imagor/pull/855
Full Changelog: https://github.com/cshum/imagor/compare/v1.9.11...v1.9.12
用户点评
登录用户的真实使用记录,可按版本和硬件筛选还没有人点评。第一个说出真实体验的人最有价值。
imagor
imagor is a fast, secure image processing server and Go library.
imagor uses one of the most efficient image processing libraries, libvips with the Go binding vipsgen. It is typically 4-8x faster than using the quickest ImageMagick settings. imagor implements libvips streaming to facilitate parallel processing pipelines and achieve high network throughput. Benchmarks show imagor is one of the fastest image processing servers.
imagor supports a wide range of image-processing use cases, available as a HTTP server with first-class Docker support. It adopts the thumbor URL syntax representing a high-performance drop-in replacement.
imagor is built with speed, security and extensibility in mind. Alongside there is imagorvideo bringing video thumbnail capability through ffmpeg C bindings, and Imagor MCP bringing template composition to agents.
Quick Start
docker run -p 8000:8000 shumc/imagor -imagor-unsafe -imagor-auto-webp
Original images:
https://raw.githubusercontent.com/cshum/imagor/master/testdata/gopher.png
https://raw.githubusercontent.com/cshum/imagor/master/testdata/dancing-banana.gif
https://raw.githubusercontent.com/cshum/imagor/master/testdata/gopher-front.png
Try out the following image URLs:
http://localhost:8000/unsafe/fit-in/200x200/filters:fill(white)/https://raw.githubusercontent.com/cshum/imagor/master/testdata/gopher.png
http://localhost:8000/unsafe/200x200/smart/filters:fill(white):format(jpeg):quality(80)/https://raw.githubusercontent.com/cshum/imagor/master/testdata/gopher.png
http://localhost:8000/unsafe/fit-in/-180x180/10x10/filters:hue(290):saturation(100):fill(yellow)/raw.githubusercontent.com/cshum/imagor/master/testdata/gopher.png
http://localhost:8000/unsafe/30x40:100x150/filters:fill(cyan)/raw.githubusercontent.com/cshum/imagor/master/testdata/dancing-banana.gif
http://localhost:8000/unsafe/fit-in/200x150/filters:fill(yellow):watermark(raw.githubusercontent.com/cshum/imagor/master/testdata/gopher-front.png,repeat,bottom,0,40,40)/raw.githubusercontent.com/cshum/imagor/master/testdata/dancing-banana.gif
Documentation
Full documentation is available at docs.imagor.net.
Templates
Imagor MCP — an MCP server and a skill that add a template layer to imagor. An agent writes a layout: OG cards, product grids, quotes, testimonials, and it compiles to one imagor URL.
Consulting
If your team is running imagor in production and needs help with architecture, custom imagor components, multi-tenant setups, loaders/storage backends, migration planning, or performance tuning, commercial support and consulting are available. See the Consulting page for details.
For bugs and feature requests, please continue to use GitHub Issues. For consulting inquiries, contact [email protected].
