Self-hosters who want instruction-based photo editing get it free under Apache 2.0, provided they have around 30GB of graphics memory.
Kenning's take
What it does
FireRed-Image-Edit is a free, open image editing model that changes photos from written instructions, such as swapping styles, adding text, virtual try-on, makeup and restoring old photos. Version 1.1 improved portrait consistency and combining many elements in one image. Weights, training code and a benchmark are published under Apache 2.0, and running it locally needs about 30GB of graphics memory with the lighter script.
- Instruction-based photo editing
- Identity consistency for people
- Combines 10+ elements in one image
- Portrait makeup styles
- ComfyUI workflow and LoRA training
- Apache 2.0 licence
Who it suits. Developers and technical creators who want a self-hosted image editing model.
Key facts
- Category
- AI models
- Made by
- Xiaohongshu (FireRed team)
- Released
- 14 February 2026
- How to get it
- Download from Hugging Face or ModelScope; code on GitHub; online demo linked from the repo
- Works on
- open weights, Hugging Face, ModelScope, ComfyUI
- In the UK
- Yes
- Free plan
- open weights, Apache 2.0
FAQ
When was FireRed-Image-Edit released?
Version 1.0 weights came out on 14 February 2026, and version 1.1 on 3 March 2026.
Is FireRed-Image-Edit free?
Yes. It is open source under the Apache 2.0 licence.
Who made FireRed-Image-Edit?
A team at Xiaohongshu, according to the GitHub repository.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools

ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.

Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.

Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.