Recommend VAE:<a target="_blank" rel="ugc" href="https://huggingface.co/Owen777/UltraFlux-v1/blob/main/vae/diffusion_pytorch_model.safetensors">https://huggingface.co/Owen777/UltraFlux-v1/blob/main/vae/diffusion_pytorch_model.safetensors</a>Online generation<a target="_blank" rel="ugc" href="https://tensor.art/models/952032833109521879/2127-Z-Image-Asian-Utopian-Turbo-v2.0-FF">https://tensor.art/models/952032833109521879/2127-Z-Image-Asian-Utopian-Turbo-v2.0-FF</a><hr /><h2 id="v3.6-baseturbo-ffv">v3.6 - Base/Turbo - FFV</h2>使用 v2.0-FF 加入全新的訓練集，調整了光澤、臉孔、體態。We used v2.0-FF to add a brand new training set and adjusted the gloss, face, and body shape.バージョン2.0-FFを使用して、全く新しいトレーニングセットを追加し、光沢、顔、体型を調整しました。저희는 v2.0-FF를 사용하여 완전히 새로운 학습 데이터 세트를 추가하고 광택, 얼굴 및 몸매를 조정했습니다.<hr /><h2 id="v3.0-turbo-ffdpo">v3.0 - Turbo - FFDPO</h2>使用 v2.0 fp32 版本，並重新融合 Lokr 與 <a target="_blank" rel="ugc" href="https://huggingface.co/F16/z-image-turbo-flow-dpo">FDPO Lora</a>。轉換為 bf16 與 Q8_0 gguf。Use version v2.0 fp32 and merge Lokr and <a target="_blank" rel="ugc" href="https://huggingface.co/F16/z-image-turbo-flow-dpo">FDPO Lora</a>. Convert to bf16 and Q8_0 gguf.バージョンv2.0 fp32を使用し、Lokrと <a target="_blank" rel="ugc" href="https://huggingface.co/F16/z-image-turbo-flow-dpo">FDPO Lora </a>を再マージします。bf16とQ8_0 ggufに変換します。버전 v2.0 fp32를 사용하고 Lokr 및 <a target="_blank" rel="ugc" href="https://huggingface.co/F16/z-image-turbo-flow-dpo">FDPO Lora </a>를 다시 병합합니다. bf16 및 Q8_0 gguf로 변환합니다.<hr /><h2 id="v2.0-ff-v2.0-base">v2.0 - FF / v2.0 - Base</h2>FF = 全片幅，全精度使用 2.4MP 高解析度資料集訓練 fp32, dim 64 Lora，並重新調整 v1.5 - ZSVD 融合配重，重新製作為 fp32 全精度版本，轉換為 bf16, GGUF。 FF = Full Frame, Full PrecisionThe fp32, dim 64 Lora dataset was trained using a 2.4MP high-resolution dataset, and the v1.5 - ZSVD fusion weights were readjusted to create a new fp32 full-precision version, which was then converted to bf16 and GGUF.FF = フルフレーム、フル精度fp32、dim 64のLoraデータセットは、2.4MPの高解像度データセットを用いて学習され、v1.5 - ZSVD融合重みが再調整されて新しいfp32フル精度バージョンが作成され、その後bf16とGGUFに変換されました。FF = 풀 프레임, 풀 정확도fp32, dim 64 의 Lora 데이터 세트는 2.4MP의 고해상도 데이터 세트를 사용하여 학습되었으며 v1.5 - ZSVD 융합 가중치가 재조정되어 새로운 fp32 풀 정확도 버전이 생성된 다음 bf16 및 GGUF 로 변환되었습니다.<hr /><h2 id="v1.5-zsvd">v1.5 - ZSVD</h2>額外使用 1000 張高畫質訓練 dim 128 Lora 並 SVD 融合 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora v3.78</a> 與原版 Z Image Turbo bf16。An additional 1000 high-quality images were used to train dim 128 Lora and SVD fused with <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora v3.78</a> and the original Z Image Turbo bf16.追加の 1000 枚の高品質画像を使用して、 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora v3.78</a> とオリジナルの Z Image Turbo bf16 を融合した dim 128 Lora と SVD をトレーニングしました。추가로 1000개의 고품질 이미지를 사용하여 dim 128 Lora 와 SVD를 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora v3.78</a> 및 기존 Z Image Turbo bf16과 융합하여 학습시켰습니다.<hr /><h2 id="v1.0-turbo">v1.0 - Turbo</h2>這是使用了 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora</a> v3.78 與原版 Z Image Turbo bf16，另外加入了兩組 10k 資料集（我擁有所有使用權，其實，都是我自行拍攝的風景照），使用 <code>gradient_accumulation: 4</code> 額外訓練的 Lora 來調配權重，並根據 AI-Toolkit 的訓練結果中，將 DiT 根據 30 個不同的 Blocks 各別調整融合成的模型。 如果可以的話，請幫我返圖。This model was created using <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora</a> v3.78 and the original Z Image Turbo bf16, with two additional 10k datasets (I own all rights to these datasets; they are actually landscape photos I took myself). It uses an additional Lora dataset trained with <code>gradient_accumulation: 4</code> to adjust the weights, and the DiT algorithm is adjusted and fused based on the training results from the AI-Toolkit, tailored to 30 different blocks.Feel free to upload the images you generated.このモデルは、 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora</a> v3.68 とオリジナルの Z Image Turbo bf16 を使用し、さらに2つの10kデータセット（すべて私が著作権を所有しており、実際に私が撮影した風景写真です）を追加しました。重み調整のため <code>gradient_accumulation: 4</code> で学習した Lora モデルを追加し、DiT アルゴリズムは AI-Toolkit の学習結果に基づいて調整・融合され、30 個の異なるブロックに合わせて調整されています。可能であれば、生成した画像をアップロードしてください。이 모델은 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2185778/z-image-turbo-asian-mix-lora">[Z Image Turbo] Asian Mix Lora</a> v3.78 과 원래 Z Image Turbo bf16 을 사용했으며, 두 개의 10k 데이터 세트 (모두 내가 저작권을 소유하고 실제로 내가 찍은 풍경 사진입니다) 을 추가했습니다. 가중치 조정을 위해, <code>gradient_accumulation: 4</code> 에서 학습한 Lora 모델을 추가하고, DiT 알고리즘은 AI-Toolkit 의 학습 결과에 근거해 조정·융합되어, 30 개의 다른 블록에 맞추어 조정되고 있습니다.가능하면 생성된 이미지를 업로드하세요.

2127 - Z Image Asian Utopian - Turbo/Base BF16/GGUF

The fp8 has too much loss; GGUF Q8_0 will be better.

2127ZImageAsianUtopian_v20Base.safetensors

2127ZImageAsianUtopian_v20Base.gguf

v2.0 - Base

v3.6 - Turbo - FFV

v3.6 - Base - FFV

v3.0 - Turbo - FFDPO

v2.0 - FF

v1.5 - ZSVD

v1.0 - Turbo

Files with SHA256: e39ed725b434cd862ea271a988f8920d18a9c45c7c890f780dd19d0fd8d7680d