From 35dca4021682ca7112bf5226703d51b07b6724d5 Mon Sep 17 00:00:00 2001 From: ComfyUI Wiki Date: Fri, 1 Aug 2025 12:53:32 +0800 Subject: [PATCH] Update model links --- tutorials/flux/flux1-krea-dev.mdx | 13 +++++++++---- zh-CN/tutorials/flux/flux1-krea-dev.mdx | 17 ++++++++++------- 2 files changed, 19 insertions(+), 11 deletions(-) diff --git a/tutorials/flux/flux1-krea-dev.mdx b/tutorials/flux/flux1-krea-dev.mdx index 05c76982..3542a2ac 100644 --- a/tutorials/flux/flux1-krea-dev.mdx +++ b/tutorials/flux/flux1-krea-dev.mdx @@ -38,7 +38,12 @@ Download the image or JSON below and drag it into ComfyUI to load the correspond Please download the following model files: **Diffusion model** +- [flux1-krea-dev_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/FLUX.1-Krea-dev_ComfyUI/blob/main/split_files/diffusion_models/flux1-krea-dev_fp8_scaled.safetensors) + +If you want to pursue higher quality and have enough VRAM, you can try the original model weights + - [flux1-krea-dev.safetensors](https://huggingface.co/black-forest-labs/FLUX.1-Krea-dev/resolve/main/flux1-krea-dev.safetensors) + The `flux1-dev.safetensors` file requires agreeing to the [black-forest-labs/FLUX.1-Krea-dev](https://huggingface.co/black-forest-labs/FLUX.1-Krea-dev/) agreement before downloading via browser. @@ -58,7 +63,7 @@ File save location: ComfyUI/ ├── models/ │ ├── diffusion_models/ -│ │ └── flux1-krea-dev.safetensors +│ │ └── flux1-krea-dev_fp8_scaled.safetensors or flux1-krea-dev.safetensors │ ├── text_encoders/ │ │ ├── clip_l.safetensors │ │ └── t5xxl_fp16.safetensors or t5xxl_fp8_e4m3fn.safetensors @@ -77,9 +82,9 @@ Please refer to the image below to ensure all model files have been loaded corre ![ComfyUI Flux Krea Dev Workflow](/images/tutorial/flux/flux_1_krea_dev_guide.jpg) -1. Ensure that `flux1-krea-dev.safetensors` is loaded in the `Load Diffusion Model` node - - `weight_dtype` has been set to `fp8_e4m3fn_fast` for low VRAM users - - If you have sufficient VRAM like 24GB, you can set `weight_dtype` to `default` for better quality +1. Ensure that `flux1-krea-dev_fp8_scaled.safetensors` or `flux1-krea-dev.safetensors` is loaded in the `Load Diffusion Model` node + - `flux1-krea-dev_fp8_scaled.safetensors` is recommended for low VRAM users + - `flux1-krea-dev.safetensors` is the original weights, if you have enough VRAM like 24GB you can use it for better quality 2. Ensure the following models are loaded in the `DualCLIPLoader` node: - clip_name1: t5xxl_fp16.safetensors or t5xxl_fp8_e4m3fn.safetensors - clip_name2: clip_l.safetensors diff --git a/zh-CN/tutorials/flux/flux1-krea-dev.mdx b/zh-CN/tutorials/flux/flux1-krea-dev.mdx index d4bf69f4..0bd2b708 100644 --- a/zh-CN/tutorials/flux/flux1-krea-dev.mdx +++ b/zh-CN/tutorials/flux/flux1-krea-dev.mdx @@ -33,10 +33,13 @@ import UpdateReminder from '/snippets/zh/tutorials/update-reminder.mdx'

下载 JSON 格式工作流

-#### 2. 手动安装模型 +#### 2. 模型链接 -请下载下面的模型文件： -**Diffusion model** +**Diffusion model** 下面两个模型选择其中一个版本即可 + +- [flux1-krea-dev_fp8_scaled.safetensors](https://huggingface.co/Comfy-Org/FLUX.1-Krea-dev_ComfyUI/blob/main/split_files/diffusion_models/flux1-krea-dev_fp8_scaled.safetensors) + +下面这个版本是原始权重，如果你追求更高质量有足够的显存，可以尝试这个版本 - [flux1-krea-dev.safetensors](https://huggingface.co/black-forest-labs/FLUX.1-Krea-dev/resolve/main/flux1-krea-dev.safetensors) @@ -58,7 +61,7 @@ import UpdateReminder from '/snippets/zh/tutorials/update-reminder.mdx' ComfyUI/ ├── models/ │ ├── diffusion_models/ -│ │ └── flux1-krea-dev.safetensors +│ │ └── flux1-krea-dev_fp8_scaled.safetensors 或 flux1-krea-dev.safetensors │ ├── text_encoders/ │ │ ├── clip_l.safetensors │ │ └── t5xxl_fp16.safetensors 或 t5xxl_fp8_e4m3fn.safetensors @@ -77,9 +80,9 @@ ComfyUI/ ![ComfyUI Flux Krea Dev工作流](/images/tutorial/flux/flux_1_krea_dev_guide.jpg) -1. 确保在`Load Diffusion Model`节点加载了`flux1-krea-dev.safetensors` - - `weight_dtype`已为低显存用户设置成了 `fp8_e4m3fn_fast` - - 如果你有足够的显存如 24GB 可以将 `weight_dtype` 设置为 `default` 以获取更好的质量 +1. 确保在`Load Diffusion Model`节点加载了`flux1-krea-dev_fp8_scaled.safetensors` 或 `flux1-krea-dev.safetensors` + - `flux1-krea-dev_fp8_scaled.safetensors` 推荐低显存用户使用 + - `flux1-krea-dev.safetensors` 如果你有足够的显存如 24GB， 你可以尝试这个版本以追求更好的质量 2. 确保在`DualCLIPLoader`节点中下面的模型已加载： - clip_name1: t5xxl_fp16.safetensors 或 t5xxl_fp8_e4m3fn.safetensors - clip_name2: clip_l.safetensors