---
title: libwaifu-krea2-turbo
canonical_url: "https://www.modelscope.cn/models/ling0322/libwaifu-krea2-turbo"
md_url: "https://www.modelscope.cn/models/ling0322/libwaifu-krea2-turbo.md"
repository: ling0322/libwaifu-krea2-turbo
chinese_name: "Krea 2 Turbo (libwaifu 包)"
last_updated: 2026-09-24
license: other
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - krea/Krea-2-Turbo
base_model_relation: finetune
parameters: 32.5B
tensor_type:
  - F16
  - F32
  - F8_E4M3
library_name:
  - safetensors
language:
  - en
downloads: 19
stars: 0
tags:
  - libwaifu
  - krea2
---

# libwaifu-krea2-turbo

> libwaifu-krea2-turbo - ling0322 在 ModelScope 开源的模型。Krea 2 Turbo — the libwaifu package

ling0322/libwaifu-krea2-turbo 是 ModelScope 魔搭社区上的 32.5B 参数text-to-image-synthesis模型，采用 other 许可，基于 krea/Krea-2-Turbo 构建。

- **Repository**: ling0322/libwaifu-krea2-turbo
- **License**: other
- **Tasks**: text-to-image-synthesis
- **Parameters**: 32.5B
- **Base model**: krea/Krea-2-Turbo
- **Tags**: libwaifu, krea2
- **Downloads**: 19
- **Stars**: 0
- **Last updated**: 2026-09-24

Source: https://www.modelscope.cn/models/ling0322/libwaifu-krea2-turbo

---

# Krea 2 Turbo — the libwaifu package

**Krea 2 Turbo**, converted to the package format [libwaifu](https://github.com/ling0322/libwaifu)
reads. The same weights drawing the same pictures; a different file layout and a narrower dtype.

> **By downloading these files you agree to the [Krea 2 Community License
> Agreement](https://krea.ai/krea-2-licensing)**, a copy of which is distributed here as
> `LICENSE.pdf`, and to the [Acceptable Use Policy](https://www.krea.ai/krea-2-use-policy).
> Commercial use is permitted only under the revenue threshold in §2.3 of that agreement, and §4.2
> requires a deployment of these weights to carry content filtering.
>
> **This is a modified copy** — see `NOTICE` for what was changed — and it is **not an official
> Krea product and is not endorsed by Krea**. The original weights are at
> [krea/Krea-2-Turbo](https://huggingface.co/krea/Krea-2-Turbo).

## What is here

| package | on disk | on the card | what it is |
|---|---|---|---|
| `krea2-turbo.yaml` and its parts | 33.8 GB | ~34 GB | float16 throughout |
| `krea2-turbo-fp8.yaml` and its parts | 17.3 GB | ~18 GB | the matrices as E4M3 with one scale for the whole tensor, everything else float16 |

Either one carries the whole model: the denoiser, the language half of the Qwen3-VL text encoder,
the Qwen-Image autoencoder, and the tokenizer. Nothing else has to be fetched to draw.

## Drawing with it

```bash
waifu draw -m krea2-turbo.yaml
```

or from Rust:

```rust
let manifest = Manifest::open("krea2-turbo.yaml")?;
let model = Krea2::from_manifest(Device::Cuda, Residency::Device, &manifest)?;
let image = model.generate("a red fox sitting in fresh snow", &options)?;
```

It is distilled for **eight steps at no guidance**. The package says so in its `suggested:` block,
and the thirty steps at guidance five that SDXL likes give a burnt picture here for four times the
work.

## How close it is to the original

Measured against diffusers' `Krea2Pipeline` on the same weights, with every input taken off one
real eight-step trajectory rather than from `torch.randn`:

| | float16 package | fp8 package |
|---|---|---|
| the twelve tapped encoder states | 1.06e-2 | 4.93e-2 |
| one denoising step | 3.60e-2 | 5.02e-2 |
| the autoencoder's decode | 1.72e-3 | the same — it is not quantized |
| eight steps, end to end | 9.94e-2 | 3.51e-1 |

For the float16 package those are what the two *float types* disagree by and little else: the
reference's own float16-against-bfloat16 run moves the same denoising step by 3.51e-2, and at full
width on the processor the autoencoder agrees with the reference to 1.9e-6.

**The fp8 package is not free.** It costs about five times the error in the text encoder, and its
eight-step trajectory ends somewhere measurably different — the same scene, drawn a little
differently. Take it when the card is the constraint, not when the picture is.

`docs/krea2.md` in the libwaifu repository has the full table, the control each number is read
against, and what is easy to get wrong in this architecture.

## Attribution

Krea 2 is licensed under the Krea 2 Community License Agreement. For more information, visit
https://krea.ai/krea-2-licensing.