---
license: other
license_name: krea-2-community-license
license_link: https://krea.ai/krea-2-licensing
base_model: krea/Krea-2-Raw
tags:
- image-editing
- lora
- comfyui
- krea-2
---

# Krea 2 Identity Edit

> **v1.1 (recommended): `krea2_identity_edit_v1_1.safetensors`** — substantially
> improved face likeness and image fidelity, much stronger edit locality
> (camera, pose, and untouched elements stay fixed far more reliably), better
> two-person identity separation, more reliable object remove/replace,
> better compound outfit-change compliance, corrected reference geometry
> handling. One honest regression: *person*-replacement ("replace the
> woman with an orangutan") is currently weaker than v1 — keep v1 for that
> use case until v1.2. No high-resolution adaptation pass yet: at high
> resolutions (especially two-person edits) identities can bleed together —
> prefer ~1–1.5MP and upscale. v1 remains available for workflow
> reproducibility. **Low-VRAM variants:** `..._v1_1_r128.safetensors` (0.91GB)
> and `..._v1_1_r64.safetensors` (0.46GB) — SVD rank-reduced from v1.1,
> near-identical quality.

**Instruction-based, identity-preserving image editing for Krea 2** (12.9B
single-stream MMDiT). Give it an image and a plain-language instruction; it
edits while preserving what you didn't ask to change — including the person.

*An unofficial community fine-tune of [Krea 2 Raw](https://huggingface.co/krea/Krea-2-Raw).
Not an official Krea product; not affiliated with or endorsed by Krea.ai, Inc.*

**Requires the [ComfyUI-Krea2Edit node pack](https://github.com/lbouaraba/comfyui-krea2edit)** — the LoRA is
trained with dual conditioning (in-context VAE tokens + image-grounded Qwen3-VL
encoding) that stock nodes don't provide. Two ready-made workflows ship with it.

## What it does

- **Person re-staging with likeness:** "create a photo of this person at a
  night market" — same face, same outfit down to individual moles and marks,
  fully relit to the new scene. New camera angles and poses included.
- **Local edits:** recolor, add/remove/replace objects, attribute and outfit
  changes, with near-pixel preservation of the rest of the frame.
- **Replace-with-reference:** "replace the woman with a big orangutan" — the
  replace verb is trained, locality holds.
- **Full-image restyles:** global style with preserved composition.
- **Two-input edits:** scene + person as separate references — "create a photo
  of this man next to the tractor." **Input order matters and is fixed: the
  scene is always image 1 (`source_latent`/`image`), the person is always
  image 2 (`source_latent_b`/`image_b`).** Swapping them sharply degrades
  results (this matches the training layout).
- **Composes with your LoRAs:** character/body/style LoRAs stack on top and
  steer the prior — something closed editors structurally can't offer.

## Recommended settings

| Task type | Model | Steps | CFG |
|---|---|---|---|
| Most edits (add, recolor, restyle, re-stage) | Turbo | 8–12 | 1.0 |
| Removals / large deletions | Raw | 20 | 3.0 |

- **Match the output aspect ratio to the source image.** Training pairs are
  same-size; AR mismatch degrades preservation (edits may apply to only part
  of the frame).
- **Generate at ≤2MP** (source bleed / duplication above). **For v1.1
  two-person edits, prefer ~1–1.5MP** — at higher resolutions the two
  identities may blend together; generate lower and upscale instead.
- **Step count is a mild dial too:** fewer steps (8) favor composition
  adherence, more (12) favor face detail; ~10 is a good balance.
- **`grounding_px` is a real dial.** Lower values = stronger edit adherence
  and more uniform scene changes; higher = stronger identity/likeness.
  v1.1's trained range is 384–768 (768 default); 1024 often still works
  nicely. **If you get duplicated/split compositions ("double pictures"),
  lower `grounding_px` — running far above the trained range is the most
  common cause.** (v1's trained range was 512–1536.)
- At CFG > 1, ground the negative too (empty prompt + same image).
- LoRA strength 1.0.

## Known limitations (honest list)

- **Likeness is texture-faithful, proportion-conservative.** Moles, skin
  character, hair, and lighting adapt beautifully; strongly distinctive facial
  *geometry* (unusual nose, eye spacing, face length) regresses toward typical
  proportions. People whose identity lives in texture and structure transfer
  best; geometry-defined faces read as a "close relative."
- **Two-person inputs keep outfits distinct but faces drift toward each
  other.** Workaround that works today: chain single-ref inserts (place person
  A, then a second edit pass adding person B from their reference).
- **Removal works but is not yet reliable** — always use the Raw/CFG 3 recipe;
  expect occasional re-renders instead of deletions.
- **Outfit swaps are hit-or-miss** — changing what a person wears sometimes
  works cleanly and sometimes doesn't apply; reroll or rephrase.
- **Local edits aren't always perfectly local** — add/remove/replace operations
  can sometimes alter other parts of the frame or shift the overall color
  grade (substantially improved in v1.1). If preservation matters, compare
  against the source and reroll.
- Highly unusual visual content (extravagant hairstyles, extreme body types)
  can drift toward the base prior — a subject LoRA stacked on top fixes this.

## License

The LoRA weights are a **Derivative Model of Krea 2** and are distributed under
the **[Krea 2 Community License Agreement](LICENSE.pdf)** (see also `NOTICE`).
Key points for users: commercial use is permitted under the license's revenue
threshold (§2.3, currently <$1M/yr — above that, contact Krea for an enterprise
license); deployments must implement reasonable content moderation (§4.2); AI
disclosure obligations apply where required (§4.3). This repository modifies
the Krea Model as permitted by §3; it is not endorsed by Krea.

Research/portfolio release by a self-funded hobbyist.

## Showcase

All reference people below are themselves AI-generated — no real likenesses.
Prompts are embedded in each image.

### v1.1 (recommended)

![Two references, one scene — the v1.1 flagship](showcase/release_v11_thumb.png)
![v1 vs v1.1 — same scene, same references](showcase/release_v11_9.png)

![Scene + person insert — scene must be image 1](showcase/release_v11_1.png)
![v1 vs v1.1 — instruction adherence and likeness](showcase/release_v11_2.png)
![Outfit replacement — v1.1 executes compound garment instructions](showcase/release_v11_3.png)
![Object removal — v1.1 removes and resolves the pose; v1 ignores the instruction](showcase/release_v11_4.png)
![Object replacement — v1.1 replaces in place; v1 adds instead](showcase/release_v11_5.png)
![grounding_px sweep — high values can duplicate the subject; lower to fix](showcase/release_v11_6.png)
![Recolor — chained edit of a previous output, everything else preserved](showcase/release_v11_7.png)
![Restoration — near out-of-distribution: a single restoration sample exists in the training data](showcase/release_v11_8.png)

### v1 gallery

![Two-reference composition](showcase/release_1.png)
![Outfit swap + scene + weather](showcase/release_2.png)
![Person re-staging with relight](showcase/release_3.png)
![New camera angle](showcase/release_4.png)
![Replace with preservation](showcase/release_5.png)
![Full-image restyle](showcase/release_6.png)
![Object addition](showcase/release_7.png)