v3.3.1

mudler/LocalAIv3.3.1Aug 1, 2025by mudler

AI Summary

Introduces Flux Kontext and Flux Krea support for image editing. Adds support for reference images in stablediffusion-ggml and allows loading LoRAs. Includes a breaking change regarding Intel GPU image naming.

Key Highlights

  • Support for Flux Kontext image editing.
  • Added Flux Krea model.
  • Stablediffusion-ggml now supports reference images and LoRA loading.
  • Intel GPU Docker images renamed to a single `latest-gpu-intel` tag.

Breaking Changes

  • Intel GPU container images renamed from `latest-gpu-intel-f32/f16` to `latest-gpu-intel`.

New Features

  • Flux Kontext support.
  • Flux Krea support.
  • Reference images support in diffusers.
  • LoRA loading support in stablediffusion-ggml.

Full Release Notes

<!-- Release notes generated using configuration in .github/release.yml at master -->

This is a minor release, however we have addressed some important bug regarding Intel-GPU Images, and we have changed naming of the container images.

This release also adds support for Flux Kontext and Flux krea!

## :warning:  Breaking change

Intel GPU images has been renamed from `latest-gpu-intel-f32` and `latest-gpu-intel-f16` to a single one, `latest-gpu-intel`, for example:

```bash
docker run -ti --name local-ai -p 8080:8080 --device=/dev/dri/card1 --device=/dev/dri/renderD128 localai/localai:latest-gpu-intel
```

and for AIO (All-In-One) images:

```bash
docker run -ti --name local-ai -p 8080:8080 localai/localai:latest-aio-gpu-intel
```

## :framed_picture:  Flux kontext

From this release LocalAI supports Flux Kontext and can be used to edit images via the API:

Install with:

```bash
local-ai run flux.1-kontext-dev
```

To test:

```bash
curl http://localhost:8080/v1/images/generations -H "Content-Type: application/json" -d '{
  "model": "flux.1-kontext-dev",
  "prompt": "change 'flux.cpp' to 'LocalAI'",
  "size": "256x256",
  "ref_images": [
  	"https://raw.githubusercontent.com/leejet/stable-diffusion.cpp/master/assets/flux/flux1-dev-q8_0.png"
  ]
}'
```

<img width="256" height="256" alt="b64567298114 (1)" src="https://github.com/user-attachments/assets/5f659956-9e7e-4517-998b-7b373d0c081c" />
<img width="256" height="256" alt="b641424088517" src="https://github.com/user-attachments/assets/045cb064-015b-4393-8b08-1c90c1f79005" />


## What's Changed
### Breaking Changes 🛠
* fix(intel): Set GPU vendor on Intel images and cleanup by @richiejp in https://github.com/mudler/LocalAI/pull/5945
### Exciting New Features 🎉
* feat(stablediffusion-ggml): add support to ref images (flux Kontext) by @mudler in https://github.com/mudler/LocalAI/pull/5935
### 🧠 Models
* chore(model gallery): add qwen_qwen3-30b-a3b-instruct-2507 by @mudler in https://github.com/mudler/LocalAI/pull/5936
* chore(model gallery): add arcee-ai_afm-4.5b by @mudler in https://github.com/mudler/LocalAI/pull/5938
* chore(model gallery): add qwen_qwen3-30b-a3b-thinking-2507 by @mudler in https://github.com/mudler/LocalAI/pull/5939
* chore(model gallery): add flux.1-dev-ggml-q8_0 by @mudler in https://github.com/mudler/LocalAI/pull/5947
* chore(model gallery): add flux.1-dev-ggml-abliterated-v2-q8_0 by @mudler in https://github.com/mudler/LocalAI/pull/5948
* chore(model gallery): add flux.1-krea-dev-ggml by @mudler in https://github.com/mudler/LocalAI/pull/5949
### Other Changes
* docs: :arrow_up: update docs version mudler/LocalAI by @localai-bot in https://github.com/mudler/LocalAI/pull/5929
* chore: :arrow_up: Update ggml-org/llama.cpp to `8ad7b3e65b5834e5574c2f5640056c9047b5d93b` by @localai-bot in https://github.com/mudler/LocalAI/pull/5931
* chore: :arrow_up: Update leejet/stable-diffusion.cpp to `f6b9aa1a4373e322ff12c15b8a0749e6dd6f0253` by @localai-bot in https://github.com/mudler/LocalAI/pull/5930
* chore: :arrow_up: Update ggml-org/whisper.cpp to `d0a9d8c7f8f7b91c51d77bbaa394b915f79cde6b` by @localai-bot in https://github.com/mudler/LocalAI/pull/5932
* chore: :arrow_up: Update ggml-org/llama.cpp to `aa79524c51fb014f8df17069d31d7c44b9ea6cb8` by @localai-bot in https://github.com/mudler/LocalAI/pull/5934
* chore: :arrow_up: Update ggml-org/llama.cpp to `e9192bec564780bd4313ad6524d20a0ab92797db` by @localai-bot in https://github.com/mudler/LocalAI/pull/5940
* chore: :arrow_up: Update ggml-org/whisper.cpp to `f7502dca872866a310fe69d30b163fa87d256319` by @localai-bot in https://github.com/mudler/LocalAI/pull/5941
* chore: update swagger by @mudler in https://github.com/mudler/LocalAI/pull/5946
* feat(stablediffusion-ggml): allow to load loras by @mudler in https://github.com/mudler/LocalAI/pull/5943
* chore(capability): improve messages by @mudler in https://github.com/mudler/LocalAI/pull/5944
* feat(swagger): update swagger by @localai-bot in https://github.com/mudler/LocalAI/pull/5950
* chore: :arrow_up: Update ggml-org/llama.cpp to `daf2dd788066b8b239cb7f68210e090c2124c199` by @localai-bot in https://github.com/mudler/LocalAI/pull/5951


**Full Changelog**: https://github.com/mudler/LocalAI/compare/v3.3.0...v3.3.1