---
title: Kroma-Quantizations
canonical_url: "https://www.modelscope.cn/models/Clybius/Kroma-Quantizations"
md_url: "https://www.modelscope.cn/models/Clybius/Kroma-Quantizations.md"
repository: Clybius/Kroma-Quantizations
last_updated: 2026-11-03
license: other
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - Krea/Krea-2
base_model_relation: finetune
parameters: 54.2B
tensor_type:
  - F32
  - U8
  - F8_E4M3
  - BF16
  - F16
  - I8
library_name:
  - lora
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 20
stars: 0
tags:
  - krea2
  - krea
  - text-to-image
  - comfyui
---

# Kroma-Quantizations

> Kroma-Quantizations - Clybius 在 ModelScope 开源的模型。Kroma v0.2 — Krea 2 fine-tune (full model, quantized to ComfyUI formats)

Clybius/Kroma-Quantizations 是 ModelScope 魔搭社区上的 54.2B 参数text-to-image-synthesis模型，采用 other 许可，基于 Krea/Krea-2 构建。

- **Repository**: Clybius/Kroma-Quantizations
- **License**: other
- **Tasks**: text-to-image-synthesis
- **Parameters**: 54.2B
- **Base model**: Krea/Krea-2
- **Tags**: krea2, krea, text-to-image, comfyui
- **Downloads**: 20
- **Stars**: 0
- **Last updated**: 2026-11-03

Source: https://www.modelscope.cn/models/Clybius/Kroma-Quantizations

---

# Kroma v0.2 — Krea 2 fine-tune (full model, **quantized to ComfyUI formats**)

**[Kroma](https://huggingface.co/lodestones/Kroma)** is a fine-tune of **Krea 2** by Lodestone-Rock. This repo contains quantized .safetensors files in various ComfyUI formats of the Kroma model.

---

## Running in ComfyUI

### 1. Requirements
- A recent **ComfyUI** with native **Krea 2** support (update to the latest master).
- The **Krea 2** text encoder and VAE (the diffusion model is this repo — no separate
  base checkpoint needed).

### 2. Files to download
Place these in your ComfyUI folders:

| What | Where | Notes |
|------|-------|-------|
| `kroma-v0.2-turbo.safetensors` | `ComfyUI/models/diffusion_models/` | this repo |
| Krea 2 text encoder (Qwen3-VL) | `ComfyUI/models/text_encoders/` | load with CLIPLoader type `krea2` |
| Krea 2 VAE | `ComfyUI/models/vae/` | |

### 3. Sampling settings
The Krea 2 Turbo delta is merged in, so sample with **Turbo settings**:

| Setting | Value |
|---------|-------|
| Steps | **8-12** |
| Guidance (CFG) | **1.0 – 1.5** |
| Shift (mu) | **1.15** |

> These are the settings the Turbo checkpoint was distilled for; keep them for best
> results.

### 4. Workflow
1. **Load Diffusion Model** → select `kroma-v0.2-turbo.safetensors`.
2. **Load CLIP** → the Krea 2 text encoder, set type to **`krea2`**.
3. **Load VAE** → the Krea 2 VAE.
4. **KSampler** → steps / CFG / shift per the table above.
5. Prompt and generate.

**Minimal node chain:**
```
Load Diffusion Model (kroma-v0.2) ─► KSampler ─► VAE Decode ─► Save Image
Load CLIP (krea2) ─────────────────► (clip)       ▲
Load VAE ───────────────────────────────────────┘
Empty Latent / prompt ────────────────────────────►
```

### 5. Tips
- If the output looks off, confirm the text encoder is loaded with type **`krea2`**
  (Krea 2 expects a 12-layer Qwen3-VL stack) and that you're on a current ComfyUI.

---

## Provenance
v0.2 was produced by continued full fine-tuning of the K2 (Krea 2) checkpoint behind
Kroma v0.1 — without the rank-256 LoRA delta compression used for the v0.1 release —
followed by merging the **Krea 2 Turbo delta** in as a **rank-512 LoRA**.

## License
**MIT** — see `LICENSE`. Note that **Krea 2** is governed by its own license; this
fine-tune does not grant any rights to the base weights.