Krea 2 Ostris Edit
Krea 2 edit (reference image) nodes for LoRAs trained with ai-toolkit. Encodes prompts + reference images through the Krea 2 Qwen3-VL text encoder and patches the model…
ComfyUI-Krea2-Ostris-Edit
Nodes for running Krea 2 edit LoRAs trained with ai-toolkit (krea2 arch with
model_kwargs.edit: true).
Installation
Clone this repo into your ComfyUI custom_nodes folder and restart ComfyUI:
cd ComfyUI/custom_nodes
git clone https://github.com/ostris/ComfyUI-Krea2-Ostris-Edit.git
No extra dependencies are required. The nodes appear under the ostris/krea2
category.
Nodes
Text Encode Krea 2 Ostris Edit
Inputs: clip, prompt, optional vae and image1..image3.
Output: CONDITIONING.
Encodes the prompt together with the reference images through the Krea 2
Qwen3-VL text encoder, using Krea's conditioning template with
Picture N: vision placeholders — the same layout used during training.
When a VAE is connected, each reference image is also VAE-encoded and attached
to the conditioning as reference latents for the model patch node.
Image sizing matches training: images fed to the Qwen3-VL encoder are downscaled (never upscaled) to fit 384x384 total pixels; reference latents to fit 1MP.
Note: the text encoder checkpoint must include the Qwen3-VL vision weights or the images cannot be encoded.
Krea 2 Ostris Edit Model Patch
Inputs: model, kv_cache (default off). Output: model.
Patches the Krea 2 model so it consumes the reference latents from the
conditioning. Each reference is appended to the image token sequence and
conditioned at timestep 0 (the index_timestep_zero reference method), and
the denoising prediction covers only the target image tokens. This node is
required because the stock Krea 2 model in ComfyUI ignores reference latents.
If the conditioning has no reference latents, the patched model behaves exactly like the stock model, so it is safe to leave in the graph.
kv_cache caches the reference tokens' attention K/V: they are precomputed in
a single t=0 pass and reused on every denoising step, so the references never
ride along in the per-step sequence (faster, especially at many steps). The
LoRA must be trained with ai-toolkit's kv_cache model kwarg for this to work
properly — leave it off for normally trained edit LoRAs.
Example wiring
Load Diffusion Model (krea2) -> Load LoRA -> Krea 2 Ostris Edit Model Patch -> KSampler
CLIPLoader (krea2) -> Text Encode Krea 2 Ostris Edit (prompt + images + VAE) -> positive
CLIPLoader (krea2) -> Text Encode Krea 2 Ostris Edit (negative prompt) -> negative