CONTAINERSIMAGESRUNBUILD
Generate images from text (txt2img) or from other images (img2img)
/opt/stable-diffusion)/opt/stable-diffusion/optimizedSD)stable-diffusion-1.4 model: [***]See the stable-diffusion-webui container for a faster implementation with a web interface.
Download the stable-diffusion-1.4 model (sd-v1-4.ckpt)
bashwget [***] -O /data/models/stable-diffusion/sd-v1-4.ckpt
Then run this in the container to generate images (by default, six 512x512 images with 50 refinement steps)
bashcd /opt/stable-diffusion && python3 scripts/txt2img.py --plms \ --ckpt /data/models/stable-diffusion/sd-v1-4.ckpt \ --outdir /data/images/stable-diffusion \ --prompt "a photograph of an astronaut riding a horse"
See here for options: [***]
For just one 512x512 image with 25 steps:
bashcd /opt/stable-diffusion && python3 scripts/txt2img.py --plms \ --n_samples 1 --n_iter 1 --ddim_steps 25 \ --ckpt /data/models/stable-diffusion/sd-v1-4.ckpt \ --outdir /data/images/stable-diffusion \ --prompt "two robots walking in the woods"
--W and --H (the default is 512x512)--seed to have the images be different (the default seed is 42)For Jetson Orin Nano and reduced memory usage:
bashcd /opt/stable-diffusion && python3 optimizedSD/optimized_txt2img.py \ --sampler plms --seed 42 \ --n_samples 1 --n_iter 1 --ddim_steps 25 \ --ckpt /data/models/stable-diffusion/sd-v1-4.ckpt \ --outdir /data/images/stable-diffusion \ --prompt "a photograph of an astronaut riding a horse"
To run all these steps from a script, see stable-diffusion/test.sh
stable-diffusion | |
|---|---|
| Builds | dustynv/stable-diffusion:r35.3.1 (2023-12-12, 6.1GB)dustynv/stable-diffusion:r35.4.1 (2023-12-15, 6.1GB) |
| Notes | disabled on JetPack 4 |
| Repository/Tag | Date | Arch | Size |
|---|---|---|---|
dustynv/stable-diffusion:r35.2.1 | 2023-12-14 | arm64 | 6.1GB |
dustynv/stable-diffusion:r35.3.1 | 2023-12-12 | arm64 | 6.1GB |
dustynv/stable-diffusion:r35.4.1 | 2023-12-15 | arm64 | 6.1GB |
Container images are compatible with other minor versions of JetPack/L4T:
• L4T R32.7 containers can run on other versions of L4T R32.7 (JetPack 4.6+)
• L4T R35.x containers can run on other versions of L4T R35.x (JetPack 5.1+)
To start the container, you can use jetson-containers run and autotag, or manually put together a docker run command:
bash# automatically pull or build a compatible container image jetson-containers run $(autotag stable-diffusion) # or explicitly specify one of the container images above jetson-containers run dustynv/stable-diffusion:r35.4.1 # or if using 'docker run' (specify image and mounts/ect) sudo docker run --runtime nvidia -it --rm --network=host dustynv/stable-diffusion:r35.4.1
jetson-containers runforwards arguments todocker runwith some defaults added (like--runtime nvidia, mounts a/datacache, and detects devices)
autotagfinds a container image that's compatible with your version of JetPack/L4T - either locally, pulled from a registry, or by building it.
To mount your own directories into the container, use the -v or --volume flags:
bashjetson-containers run -v /path/on/host:/path/in/container $(autotag stable-diffusion)
To launch the container running a command, as opposed to an interactive shell:
bashjetson-containers run $(autotag stable-diffusion) my_app --abc xyz
You can pass any options to it that you would to docker run, and it'll print out the full command that it constructs before executing it.
If you use autotag as shown above, it'll ask to build the container for you if needed. To manually build it, first do the system setup, then run:
bashjetson-containers build stable-diffusion
The dependencies from above will be built into the container, and it'll be tested during. Run it with --help for build options.
来自真实用户的反馈,见证轩辕镜像的优质服务
免费版仅支持 Docker Hub 加速,不承诺可用性和速度;专业版支持更多镜像源,保证可用性和稳定速度,提供优先客服响应。
免费版仅支持 docker.io;专业版支持 docker.io、gcr.io、ghcr.io、registry.k8s.io、nvcr.io、quay.io、mcr.microsoft.com、docker.elastic.co 等。
当返回 402 Payment Required 错误时,表示流量已耗尽,需要充值流量包以恢复服务。
通常由 Docker 版本过低导致,需要升级到 20.x 或更高版本以支持 V2 协议。
先检查 Docker 版本,版本过低则升级;版本正常则验证镜像信息是否正确。
使用 docker tag 命令为镜像打上新标签,去掉域名前缀,使镜像名称更简洁。
探索更多轩辕镜像的使用方法,找到最适合您系统的配置方式
通过 Docker 登录认证访问私有仓库
在 Linux 系统配置镜像加速服务
在 Docker Desktop 配置镜像加速
Docker Compose 项目配置加速
Kubernetes 集群配置 Containerd
在宝塔面板一键配置镜像加速
Synology 群晖 NAS 配置加速
飞牛 fnOS 系统配置镜像加速
极空间 NAS 系统配置加速服务
爱快 iKuai 路由系统配置加速
绿联 NAS 系统配置镜像加速
QNAP 威联通 NAS 配置加速
Podman 容器引擎配置加速
HPC 科学计算容器配置加速
ghcr、Quay、nvcr 等镜像仓库
无需登录使用专属域名加速
需要其他帮助?请查看我们的 常见问题 或 官方QQ群: 13763429