Fork仓库后点击右上角 启动 L40 预装环境 稍等一会即可开始玩耍~ 点击 执行按钮可探索更多玩法~

torch	torchvision	torchaudio	xformers	flash-attn	SageAttention
2.7.1	0.22.1	2.7.1	0.0.31.post1	2.7.4.post1	2.2.0
2.8.0	0.23.0	2.8.0	0.0.32.post2	2.8.2	2.2.0
2.9.0	0.24.0	2.9.0	0.0.33.post1	2.8.3	2.2.0
2.9.1	0.24.1	2.9.1	0.0.33.post2	2.8.3	2.2.0

开启云原生开发环境后自动进入环境，退出直接操作快捷键 crtl + c ,默认端口8675做转发访问,通过以下命令手动启动

    bash start-ai-toolkit.sh

开启云原生开发环境后通过终端,执行下面命令启动jupyterlab环境，默认端口8687做转发访问

    bash jupyterlab.sh

开启云原生开发环境后通过终端,执行一下命令启动服务，退出直接操作快捷键 crtl + c ,默认端口8188做转发访问,通过以下命令手动启动

    bash start-comfyui.sh

类型	说训练模型类型	依赖模型	CNB平台收录	CNB收录地址
image	Chroma	lodestones/Chroma1-Base	√	https://cnb.cool/ai-models/lodestones/Chroma1-Base
	Flex.1	ostris/Flex.1-alpha	√	https://cnb.cool/ai-models/ostris/Flex.1-alpha
	Flex.2	ostris/Flex.2-preview	√	https://cnb.cool/ai-models/ostris/Flex.2-preview
	Flux.1	black-forest-labs/FLUX.1-dev	√	https://cnb.cool/ai-models/black-forest-labs/FLUX.1-dev
	Flux.2	black-forest-labs/FLUX.2-dev	√	https://cnb.cool/ai-models/black-forest-labs/FLUX.2-dev
	HiDream	HiDream-ai/HiDream-I1-Full	√	https://cnb.cool/ai-models/HiDream-ai/HiDream-I1-Full
	Lumina2	Alpha-VLLM/Lumina-Image-2.0	√	https://cnb.cool/ai-models/Alpha-VLLM/Lumina-Image-2.0
	OmniGen2	OmniGen2/OmniGen2	√	https://cnb.cool/ai-models/OmniGen2/OmniGen2
	Qwen-Image	Qwen/Qwen-Image	√	https://cnb.cool/ai-models/Qwen/Qwen-Image
	SD1.5	stable-diffusion-v1-5/stable-diffusion-v1-5	√	https://cnb.cool/ai-models/stable-diffusion-v1-5/stable-diffusion-v1-5
	SDXL	stabilityai/stable-diffusion-xl-base-1.0	√	https://cnb.cool/ai-models/stabilityai/stable-diffusion-xl-base-1.0
	Z-Image-Turbo	Tongyi-MAI/Z-Image-Turbo	√	https://cnb.cool/ai-models/Tongyi-MAI/Z-Image-Turbo
	Z-Image-Turbo	ostris/zimage_turbo_training_adapter	√	https://cnb.cool/ai-models/ostris/zimage_turbo_training_adapter
	Z-Image-De-Turbo	ostris/Z-Image-De-Turbo	√	https://cnb.cool/ai-models/ostris/Z-Image-De-Turbo
NSTRUCTION	FLUX.1-Kontext-dev	black-forest-labs/FLUX.1-Kontext-dev	√	https://cnb.cool/ai-models/black-forest-labs/FLUX.1-Kontext-dev
	HiDream-E1	HiDream-ai/HiDream-E1-1	√	https://cnb.cool/ai-models/HiDream-ai/HiDream-E1-1
	Qwen-Image-Edit	Qwen/Qwen-Image-Edit	√	https://cnb.cool/ai-models/Qwen/Qwen-Image-Edit
	Qwen-Image-Edit-2509	Qwen/Qwen-Image-Edit-2509	√	https://cnb.cool/ai-models/Qwen/Qwen-Image-Edit-2509
video	Wan2.1(1.3B)	Wan-AI/Wan2.1-T2V-1.3B-Diffusers	√	https://cnb.cool/ai-models/Wan-AI/Wan2.1-T2V-1.3B-Diffusers
	Wan2.1(14B)	Wan-AI/Wan2.1-T2V-14B-Diffusers	√	https://cnb.cool/ai-models/Wan-AI/Wan2.1-T2V-14B-Diffusers
	Wan2.1-I2V(14B-480P)	Wan-AI/Wan2.1-I2V-14B-480P-Diffuser	√	https://cnb.cool/Wan-AI/Wan2.1-I2V-14B-480P-Diffusers
	Wan2.1-I2V(14B-720P)	Wan-AI/Wan2.1-I2V-14B-720P-Diffusers	√	https://cnb.cool/Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
	Wan2.2-T2V(14B)	ai-toolkit/Wan2.2-T2V-A14B-Diffusers-bf16	√	https://cnb.cool/ai-toolkit/Wan2.2-T2V-A14B-Diffusers-bf16
	Wan2.2-I2V(14B)	ai-toolkit/Wan2.2-I2V-A14B-Diffusers-bf16	√	https://cnb.cool/ai-toolkit/Wan2.2-I2V-A14B-Diffusers-bf16
	Wan2.2-TI2V(5B)	Wan-AI/Wan2.2-TI2V-5B-Diffusers	√	https://cnb.cool/ai-models/Wan-AI/Wan2.2-TI2V-5B-Diffusers

开启云原生开发环境后通过终端,执行命令需要下载类型模型(例如下载z-image模型：下载路径脚本是/workspace/training/image/Z-Image-Turbo-download.sh)，模型放在cnb缓存路径/workspace/models_cnb_cache,训练模型时注意填写全路径

    bash /workspace/training/image/Z-Image-Turbo-download.sh

cnb平台自动下载模型详细代码实现见开源项目作者旭 ComfyUI源码修改

请搜索微信号：