<h1 id="krea-anime-v2">krea anime V2</h1>A well-balanced model that combines flexibility, precision, and a high level of detail; in terms of quality, it’s hot on the heels of Krea 2 Anime UWU v1 + Qwen Image 2.1.It is recommended to use the text encoder proposed in 1.7.8-12 stepeuler\simple (I haven't tested it on others)image resolution 1–1.3–1.5–2-3 MPhave fun)<h1 id="krea-anime-1.7">krea anime 1.7</h1>The following TE is used for inference<a target="_blank" rel="ugc" href="https://civarchive.com/models/2728378/qwen3-vl-instruct-4b-heretic-7refusal-convrot-alternate-krea-2-text-encoder-for-comfyui?modelVersionId=3066989">Qwen3 VL Instruct 4b Heretic 7refusal +convrot (alternate Krea 2 text encoder for ComfyUI) - v1.0 | Krea 2 Text Encoder | Civitai</a>It’s also attached to the card, even though the author said it wasn’t necessary to credit them. The encoder is genuinely good.As for mine... um... well, it looks like this...<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/57da4867-0a47-471d-8edf-91d50d094f63/original=true/57da4867-0a47-471d-8edf-91d50d094f63.jpeg" />Yeah, it’s still taken apart, but at least it works.<h1 id="krea-2-anime">krea 2 anime</h1><hr />The model was trained on a high-quality dataset of over 3,000 images; approximately 350 of these consist of my own drawings and AI-generated images that I edited using drawing software.<hr />Qween VL annotation + manual corrections<hr />I consider this model one of the strongest and most powerful I have ever released.The download fee is nominal. The hardware required to train the Krea 2 model on a large dataset is expensive; even though I can afford to train it locally—my PC costs around $5,000–$6,000 (and I’m currently upgrading it)—it still places a massive strain on the system. I need the funds to enable generation using this model on Civitai and possibly to purchase some cosmetics on the platform. I have previously released a large number of models completely for free, and I hope you understand my decision in this instance.<hr />I was really impressed by this model's performance—specifically its speed and flexibility. I had originally planned a large-scale fine-tuning project for Z-Image, but I changed my mind and decided to focus on Krea 2 instead. At the moment, I have a "raw" dataset of 350,000 images (about 20,000 have been processed so far; manual filtering weeds out roughly 60–70% of the material, leaving a final set of around 100,000 images), as well as a high-standard, carefully curated regularization set. Additionally, I plan to compile a supplementary set of about 20,000 images featuring game characters (primarily from HoYoverse titles) if I determine they are underrepresented in the main dataset, given that the base Krea 2 model doesn't recognize them very well—and is unfamiliar with many of them entirely.<hr />int 8 modeleuller\simple 8 stepImage resolution 1.0–1.3–1.5 MP

Krea  2 Anime_ and_CGI UWU turbo

The model has learned to generate yandere and NSFW content (though the latter was not a specific focus during training).It was trained on over 60,000 images from Danbooru (covering the 2024–2026 period), using Danbooru’s own tagging/labeling system.Additionally, the weights were fine-tuned using a set of tools I developed. Although this is an INT8 model (chosen for compactness and speed), it outperforms many BF16 models in terms of quality and detail, thanks to the weight adjustments I performed.

Files with SHA256: 9e1672e8d0ac0d7ccf44104bd33ab1155e437925b56c564b8a0ed7c1e43449b5