proof guide.en-gb.md

Author: Jessica41

pages/public_cloud/ai_machine_learning/deploy_tuto_19_flux/guide.en-gb.md

@@ -1,7 +1,7 @@
 ---
 title: AI Deploy - Tutorial - Deploy FLUX Text-to-Image Models
 excerpt: How to deploy the FLUX image generation models using ComfyUI
-updated: 2025-07-25
+updated: 2025-07-31
 ---
 
 > [!primary]
@@ -11,9 +11,9 @@ updated: 2025-07-25
 
 ## Introduction
 
-[FLUX](https://github.com/black-forest-labs/flux) is a flexible family of generative models developed by [Black Forest Technologies](https://bfl.ai/). The **FLUX** models support a variety of tasks, including text-to-image generation, structural conditioning, and inpainting.
+[FLUX](https://github.com/black-forest-labs/flux){.external} is a flexible family of generative models developed by [Black Forest Technologies](https://bfl.ai/){.external}. The **FLUX** models support a variety of tasks, including text-to-image generation, structural conditioning, and inpainting.
 
-In this tutorial, we walk through the process of deploying **FLUX** models on **AI Deploy**. We will show how to use **FLUX** models interactively with [ComfyUI](https://github.com/comfyanonymous/ComfyUI) (a visual programming interface).
+In this tutorial, we walk through the process of deploying **FLUX** models on **AI Deploy**. We will show how to use **FLUX** models interactively with [ComfyUI](https://github.com/comfyanonymous/ComfyUI){.external} (a visual programming interface).
 
 ## Instructions
 
@@ -37,7 +37,7 @@ The following table lists the main FLUX variants and their intended use cases:
 | `FLUX.1 [dev]`              | [Text to Image](docs/text-to-image.md)                     | https://huggingface.co/black-forest-labs/FLUX.1-dev            | [FLUX.1-dev Non-Commercial License](model_licenses/LICENSE-FLUX1-dev) |
 | `FLUX.1 Kontext [dev]`      | [Image editing](docs/image-editing.md)                     | https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev    | [FLUX.1-dev Non-Commercial License](model_licenses/LICENSE-FLUX1-dev) |
 
-Full list is available on the [official repository](https://github.com/black-forest-labs/flux) for FLUX.1 models.
+Full list is available on the [official repository](https://github.com/black-forest-labs/flux){.external} for FLUX.1 models.
 
 ## Requirements
 
@@ -47,7 +47,7 @@ Before proceeding, ensure you have the following:
 - An AI Deploy Project created inside a [Public Cloud project](/links/public-cloud/public-cloud) in your OVHcloud account
 - A [user for AI Training & Object Storage](/pages/public_cloud/ai_machine_learning/gi_01_manage_users)
 - [The OVHcloud AI CLI](/pages/public_cloud/ai_machine_learning/cli_10_howto_install_cli) installed on your computer
-- [Docker](https://www.docker.com/get-started) installed on your computer, **or** access to a Debian Docker Instance, which is available on the [Public Cloud](/links/manager)
+- [Docker](https://www.docker.com/get-started){.external} installed on your computer, **or** access to a Debian Docker Instance, which is available on the [Public Cloud](/links/manager)
 - A Hugging Face account, with access to the FLUX model. You need to accept usage terms on the model Hugging Face page.
 - A Hugging Face access token (generate one under your Hugging Face account → Access Tokens). This one will be used to authenticate and download the model weights.
 
@@ -140,7 +140,7 @@ This command will:
 
 You can track the progress of your job using the following commands:
 
-```
+```sh
 ovhai job list
 ovhai job logs -f <job_id>
 ```
@@ -149,13 +149,13 @@ This will allow you to view the logs generated by your job, seeing the imports a
 
 You can then verify the presence of your files by checking your Object Storage bucket from the OVHcloud Control Panel or via CLI using the following command:
 
-```console
+```sh
 ovhai bucket object list <bucket_name>@GRA
 ```
 
 ### Build a Docker image
 
-Once the FLUX model weights are uploaded to Object Storage, the next step is to build a Docker image that packages [ComfyUI](https://github.com/comfyanonymous/ComfyUI) and its required dependencies. 
+Once the FLUX model weights are uploaded to Object Storage, the next step is to build a Docker image that packages [ComfyUI](https://github.com/comfyanonymous/ComfyUI){.external} and its required dependencies. 
 
 This image will later be deployed as an AI Deploy application, where the model files will be mounted at runtime from Object Storage. This avoids the need to embed large model weights directly into the container image.
 
@@ -241,7 +241,7 @@ exec python3 /app/ComfyUI/main.py --listen 0.0.0.0 --port 8188
 
 Then, launch one of the following commands from the created folder that contain your `Dockerfile` and the `entrypoint.sh` script to build your application image:
 
-```console
+```sh
 # Build the image using your machine's default architecture
 docker build -t flux-image:latest .
 
@@ -258,27 +258,28 @@ docker buildx build --platform linux/amd64 -f Dockerfile -t flux-image:latest .
 
 After building the image, tag and push it to a container registry. 
 
-In this example, we use the OVHcloud shared registry, available to every AI Deploy user. But you can also use other registires such as OVHcloud Managed Private Registry, Docker Hub, GitHub packages, ...
+In this example, we use the OVHcloud shared registry, available to every AI Deploy user. But you can also use other registires such as OVHcloud Managed Private Registry, Docker Hub, GitHub packages, etc.
 
 > [!warning]
-> **Warning**
+> 
 > The shared registry should only be used for testing purpose. Please consider attaching your own registry. More information about this can be found [here](/pages/public_cloud/ai_machine_learning/gi_07_manage_registry). The images pushed to this registry are for AI Tools workloads only, and will not be accessible for external uses.
+>
 
 You can find the address of your shared registry by launching this command:
 
-```bash
+```sh
 ovhai registry list
 ```
 
 Log in to the shared registry with your usual AI Platform user credentials:
 
-```bash
+```sh
 docker login -u <user> -p <password> <registry_address>
 ```
 
 Tag the compiled image and push it into your registry:
 
-```bash
+```sh
 docker tag flux-image <registry_address>/flux-image:latest
 docker push <registry_address>/flux-image:latest
 ```
@@ -293,7 +294,7 @@ Run the following command to deploy your application:
 
 > **Schnell**
 >>
->> ```console
+>> ```sh
 >> ovhai app run <registry_address>/flux-image:latest \
 >>   --name flux-schnell-app \
 >>   --gpu 1 \
@@ -310,7 +311,7 @@ Run the following command to deploy your application:
 >>
 > **Dev**
 >>
->> ```console
+>> ```sh
 >> ovhai app run <registry_address>/flux-image:latest \
 >>   --name flux-dev-app \
 >>   --gpu 1 \
@@ -326,7 +327,7 @@ Run the following command to deploy your application:
 >>
 > **Kontext Dev**
 >>
->> ```console
+>> ```sh
 >> ovhai app run <registry_address>/flux-image:latest \
 >>   --name flux-kontext-app \
 >>   --gpu 1 \
@@ -352,8 +353,9 @@ Run the following command to deploy your application:
 - `--env`: Sets environment variables used by `entrypoint.sh` to configure the FLUX model files symlinks.
 
 > [!warning]
-> **Warning**
+> 
 > Other FLUX variants may expect files in different folders. If you plan to use another variant, make sure to also update the environment variables to match this variant, and adjust the `entrypoint.sh` script if necessary to match new files and folder structures.
+>
 
 Once you launch the app, AI Deploy will execute the following phases:
 
@@ -363,38 +365,38 @@ Once you launch the app, AI Deploy will execute the following phases:
 
 To monitor your app progress and logs in real time, use:
 
-```bash
+```sh
 ovhai app logs -f <app_id>
 ```
 
 Once you see in the logs that ComfyUI has started and is listening on port `8188`, the app is ready to use. You can access the interface using the public URL provided by the platform, such as:
 
-```
+```console
 https://<app_id>.app.gra.ai.cloud.ovh.net
 ```
 
 You can retrieve it at any time using the following commands:
 
-```
+```sh
 ovhai app list
 ovhai app get <app_id>
 ```
 
 ### Run inference with ComfyUI
 
-Once inside the ComfyUI web interface, head to the official [ComfyUI FLUX examples](https://comfyanonymous.github.io/ComfyUI_examples/flux/) page. Find the image matching your deployed variant.
+Once inside the ComfyUI web interface, head to the official [ComfyUI FLUX examples](https://comfyanonymous.github.io/ComfyUI_examples/flux/){.external} page. Find the image matching your deployed variant.
 
 Then, drag and drop this image into your ComfyUI interface. This will automatically load the FLUX image workflow.
 
 You can now customize the text prompt or parameters as desired. Then, just click the `Run`{.action} button to start the image generation or editing process.
 
 Once the image is generated, you can view and download it directly from the ComfyUI output node.
 
->> ![FLUX model running on AI Deploy](images/flux-ai-deploy-demo.gif){.thumbnail}
+![FLUX model running on AI Deploy](images/flux-ai-deploy-demo.gif){.thumbnail}
 
 ## Go further
 
-If you want to deploy a different interface such as [AUTOMATIC1111]((https://github.com/AUTOMATIC1111/stable-diffusion-webui)) with Stable Diffusion XL, we have a [step-by-step guide](/pages/public_cloud/ai_machine_learning/deploy_tuto_18_gradio_stable_diffusion_webui) to deploy this popular Web UI on AI Deploy.
+If you want to deploy a different interface such as [AUTOMATIC1111](https://github.com/AUTOMATIC1111/stable-diffusion-webui){.external} with Stable Diffusion XL, we have a [step-by-step guide](/pages/public_cloud/ai_machine_learning/deploy_tuto_18_gradio_stable_diffusion_webui) to deploy this popular Web UI on AI Deploy.
 
 If you are interested in image generation concepts, you can learn how image generation networks work and train your own Generative Adversarial Network. Check out this AI Notebooks guide: [Create and train an image generation model](/pages/public_cloud/ai_machine_learning/notebook_tuto_14_image-generation-dcgan).
 
@@ -404,4 +406,4 @@ If you need training or technical assistance to implement our solutions, contact
 
 Please send us your questions, feedback and suggestions to improve the service:
 
-- On the OVHcloud [Discord server](https://discord.gg/ovhcloud)
+- On the OVHcloud [Discord server](https://discord.gg/ovhcloud)