Skip to content

Update image generation demo README and parameters docs#4243

Merged
atobiszei merged 7 commits into
mainfrom
atobisze_img_gen_demo_update
May 29, 2026
Merged

Update image generation demo README and parameters docs#4243
atobiszei merged 7 commits into
mainfrom
atobisze_img_gen_demo_update

Conversation

@atobiszei
Copy link
Copy Markdown
Collaborator

@atobiszei atobiszei commented May 26, 2026

-> removed exporting models to make demos tests faster
-> added using --cache_dir to speed up loading.
-> added example images for LoRA's
http://openvino-doc.iotg.sclab.intel.com/atobisze_img_gen_demo_update/index.html

@atobiszei atobiszei added the WIP Do not merge until resolved label May 26, 2026
@atobiszei atobiszei removed the WIP Do not merge until resolved label May 28, 2026
@atobiszei atobiszei marked this pull request as ready for review May 28, 2026 14:18
@atobiszei atobiszei requested review from Copilot and dkalinowski May 28, 2026 14:18
@atobiszei atobiszei requested a review from rasapala May 28, 2026 14:18
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR updates image generation documentation to reflect newer model deployment flows and LoRA parameter usage for OVMS demos and CLI reference.

Changes:

  • Adds --source_loras to the image generation parameters table.
  • Reworks image generation demo deployment examples for --source_model, pre-downloaded models, SDXL, and inpainting.
  • Updates LoRA demo examples with new model IDs, generation parameters, vector art example, and expected output images.

Reviewed changes

Copilot reviewed 2 out of 6 changed files in this pull request and generated 1 comment.

File Description
docs/parameters.md Documents the --source_loras image generation CLI parameter.
demos/image_generation/README.md Updates image generation deployment and request examples, including SDXL, inpainting, and multi-LoRA workflows.

Comment thread demos/image_generation/README.md Outdated
Comment thread demos/image_generation/README.md Outdated

::::

> **NOTE:** The model directory must contain a valid `graph.pbtxt` file. This file is auto-generated when using `--pull` or `--source_model`. If you downloaded the model manually (e.g., via `huggingface-cli`), you can generate it by running `ovms --pull` with the same `--model_repository_path` and `--source_model` parameters, or create it manually — see [Image Generation calculator reference](../../docs/image_generation/reference.md) for all available options.
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That is not the case anymore. It is possible to create graph in runtime with just model_path parameter

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We may just have --task and model path.

@atobiszei atobiszei merged commit 5658186 into main May 29, 2026
1 check passed
@atobiszei atobiszei deleted the atobisze_img_gen_demo_update branch May 29, 2026 09:18
atobiszei added a commit that referenced this pull request May 29, 2026
-> removed exporting models to make demos tests faster
-> added using ```--cache_dir``` to speed up loading.
-> added example images for LoRA's

Cherry-pick from main:
5658186
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants