Retro Chinese Classic Animation Style Krea2 LoRA
Note: All model files and custom nodes below must be downloaded via the client. If you do not have the client, Download >>>
This LoRA is a style adapter exclusively built for the Krea2 model family. It recreates the unique artistic texture of traditional Chinese animations from the last century, restoring soft hand-painted strokes, vintage flat colors and elegant oriental atmosphere. It works well for Chinese-style illustrations, fable animation scenes and traditional art portrait creation. It only works with Krea2 (Raw / Turbo) base model and is incompatible with Stable Diffusion, MOSS‑Image and other models.
Krea2 is an open-source text-to-image model developed by Krea AI, focusing on atmospheric aesthetic rendering. It has two variants: Krea2 Raw and Krea2 Turbo. Raw is the original base checkpoint with more sampling steps and strong plasticity, ideal for LoRA training. Turbo is a distilled fast-generation variant that can produce outputs in very few steps. LoRAs trained on Raw can be directly used for inference on Turbo. This model excels at lighting shaping, color atmosphere control and composition, suitable for illustrations, portraits and concept art. This retro Chinese classic animation LoRA is a community-made model. While retaining Krea2’s native strengths in lighting and composition, it brings out the visual features of vintage hand-drawn Chinese animations to quickly generate artworks with traditional oriental animation charm.
| Name | Size | Target Dir |
|---|
| qwen3vl_4b_fp8_scaled.safetensors | 5.24 GB | text_encoders |
| qwen3vl_8b_nvfp4.safetensors | 6.31 GB | text_encoders |
| qwen3vl_8b_w4a8.safetensors | 6.31 GB | text_encoders |
| qwen3vl_4b_bf16.safetensors | 8.88 GB | text_encoders |
| qwen3vl_8b_int8_convrot.safetensors | 9.35 GB | text_encoders |
| qwen3vl_8b_fp8_scaled.safetensors | 10.59 GB | text_encoders |
| qwen3vl_8b_bf16.safetensors | 17.53 GB | text_encoders |
| Name | Size | Target Dir |
|---|
| krea2_turbo-Q2_K.gguf | 4.89 GB | unet |
| krea2_turbo-Q3_K_M.gguf | 6.01 GB | unet |
| krea2_turbo-Q3_K_S.gguf | 6.01 GB | unet |
| krea2_turbo-Q4_0.gguf | 7.49 GB | unet |
| krea2_turbo-Q4_K_M.gguf | 7.49 GB | unet |
| krea2_turbo-Q4_K_S.gguf | 7.49 GB | unet |
| krea2_turbo-Q4_1.gguf | 8.18 GB | unet |
| krea2_turbo-Q5_0.gguf | 8.87 GB | unet |
| krea2_turbo-Q5_K_M.gguf | 8.87 GB | unet |
| krea2_turbo-Q5_K_S.gguf | 8.87 GB | unet |
| krea2_turbo-Q5_1.gguf | 9.67 GB | unet |
| krea2_turbo-Q6_K.gguf | 10.58 GB | unet |
| krea2_turbo-Q8_0.gguf | 13.71 GB | unet |
| Name | Size | Target Dir |
|---|
| qwen_image_vae.safetensors | 253.81 MB | vae |
| Name | Size | Target Dir |
|---|
| shangmeiying-krea2-lora1_c2-st1000.safetensors | 234.68 MB | loras |
| shangmeiying-krea2-lora1_c2-st2000.safetensors | 234.68 MB | loras |
