| --- |
| license: other |
| license_name: krea-2-community-license |
| license_link: https://www.krea.ai/krea-2-licensing |
| base_model: krea/Krea-2 |
| tags: |
| - text-to-image |
| - krea |
| - krea2 |
| - int8 |
| - convrot |
| - comfyui |
| - quantized |
| --- |
| |
| # Krea 2 — INT8 ConvRot |
|
|
| INT8 **ConvRot** quantized weights for **[Krea 2 (K2)](https://www.krea.ai/krea-2)**, for fast, low-VRAM |
| inference in ComfyUI via the [ComfyUI-INT8-Fast](https://github.com/BobJohnson24/ComfyUI-INT8-Fast) node. |
|
|
| > **This is a modified (quantized) version of the Krea 2 model.** It is **not** an official Krea release and is |
| > **not** endorsed by Krea. The original weights are © Krea, licensed under the |
| > [Krea 2 Community License](https://www.krea.ai/krea-2-licensing). |
|
|
| ConvRot is a near-lossless INT8 scheme (~GGUF-Q8 quality) that runs on the INT8 tensor cores of any NVIDIA GPU |
| with sufficient INT8 TOPS (RTX 30-series and up). On a 3090, INT8 is meaningfully faster than FP8 (which has no |
| tensor-core acceleration on Ampere) and roughly half the VRAM of BF16. |
|
|
| ## Models |
|
|
| | File | Precision | Size | Use | |
| |------|-----------|------|-----| |
| | `Krea2-Turbo-int8-ConvRot.safetensors` | INT8 ConvRot | 14.1 GB | 8-step distilled, fast text-to-image | |
| | `Krea2-Raw-int8-ConvRot.safetensors` | INT8 ConvRot | 14.1 GB | Undistilled base — fine-tuning / LoRA training / research | |
|
|
| Original BF16 checkpoints are ~26.6 GB each. Quantized with the K2 profile: the 28 main DiT blocks are INT8, |
| while the sensitive layers (`first`, `last`, `tmlp`, `tproj`, `txtfusion`, `txtmlp`) are kept in high precision. |
|
|
| ## Requirements |
|
|
| - **ComfyUI ≥ 0.25.0** (native Krea 2 support). |
| - **[ComfyUI-INT8-Fast](https://github.com/BobJohnson24/ComfyUI-INT8-Fast)** custom node (with a `krea2` |
| model-type profile — see note below). |
| - Text encoder: **`qwen3vl_4b_fp8_scaled.safetensors`** → `ComfyUI/models/text_encoders/` |
| (from [Comfy-Org/Qwen3-VL](https://huggingface.co/Comfy-Org/Qwen3-VL)). |
| - VAE: **`qwen_image_vae.safetensors`** → `ComfyUI/models/vae/` |
| (from [Comfy-Org/Qwen-Image_ComfyUI](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI)). |
| - Put the INT8 `.safetensors` in `ComfyUI/models/diffusion_models/`. |
|
|
| > **Note on the `krea2` profile:** Krea 2 is newer than INT8-Fast's built-in model list. These weights were |
| > produced with a small added `krea2` exclusion profile (keep `first`/`last`/`tmlp`/`tproj`/`txtfusion`/`txtmlp` |
| > in high precision). Loading the *pre-quantized* files here does not require that profile — ConvRot metadata |
| > is embedded per-layer — but reproducing the conversion does. |
|
|
| ## Usage |
|
|
| Load with **Load Diffusion Model INT8 (W8A8)** (`OTUNetLoaderW8A8`): |
| - `unet_name`: the INT8 file |
| - `on_the_fly_quantization`: **false** (already quantized) |
| - `enable_convrot`: **true** |
| - `model_type`: `krea2` |
|
|
| Then the standard K2 graph: `CLIPLoader (type: krea2)` → `VAELoader` → `CLIPTextEncode` → `KSampler` → `VAEDecode`. |
|
|
| Recommended sampler settings: |
| - **Turbo:** 8 steps, CFG 1.0, `euler` / `simple`, shift 1.15 (model default). |
| - **Raw:** ~52 steps, CFG ~3.5, `euler` / `simple` (undistilled base; mainly for training). |
|
|
| ## Quantization |
|
|
| - **Method:** INT8 ConvRot (row-wise INT8 with convolutional rotation), via ComfyUI-INT8-Fast. |
| - **Compute:** `torch._int_mm` on INT8 tensor cores; Triton kernels. |
| - **Quality:** near-lossless vs BF16/FP8 in testing (verified by generation). |
|
|
| ## License & attribution |
|
|
| Krea 2 is licensed under the **Krea 2 Community License Agreement**, |
| Copyright © Krea, Inc. All Rights Reserved. See `LICENSE` / https://www.krea.ai/krea-2-licensing. |
|
|
| These files are a **quantized derivative** of Krea 2; the Krea 2 model name is retained per the license. |
| Commercial use is permitted only for entities under $1M USD trailing-twelve-month revenue; above that an |
| Enterprise License from Krea is required. **Deployers must implement reasonable content-filtering** (e.g. |
| NudeNet, Falconsai/nsfw_image_detection, Hive, or human review) to prevent prohibited content, and disclose |
| AI-generated outputs where required. |
|
|
| **Credits:** [Krea](https://www.krea.ai) (base model) · [BobJohnson24/ComfyUI-INT8-Fast](https://github.com/BobJohnson24/ComfyUI-INT8-Fast) (INT8 ConvRot method). |
|
|
