Update April 28th Added version V4 pruned model. A model with professional RPG industry artists in mind. Those working in video games, board and tabletop games as well as concept art and book covers should get good use from this model. The new version, V4, packs in more training for creatures and RPG subjects including a more illustrative style.<a target="_blank" rel="ugc" href="https://civarchive.com/models/34192/ultra-sharp-high-contrast-tutorial-vaeandupscaler">Detailed tutorial on how I get the results in the preview images.</a> Check here if you're having trouble getting the same results. Initial generation size and VAE usage is key.You can prompt any style you need with these models, but the default aesthetic is listed for each of the models in this handy list. Different models available, check the blue tabs above the images up top:Stable Diffusion 1.5 (512) versions:<ul><li>V4 inpaint Inpainting version of V4 that's good for outpainting.</li><li>V4+VAE Same as V4 but with the added convenience of having a preset VAE baked in so you don't need to select that each time. Saves on vram usage and possible NaN errors. Speeds up workflow if that's the VAE you're going to use anyway.</li><li>V4 New aesthetic added to make it a bit more illustrative. More training for better coherency for all things RPG.</li><li>V3+VAE Same as V3 but with the added convenience of having a preset VAE baked in so you don't need to select that each time. Saves on vram usage and possible NaN errors. Speeds up workflow if that's the VAE you're going to use anyway.</li><li>V3 Stronger painterly style. High contrast and sharpness. Even more RPG knowledge.</li><li>V3 inpaint Inpainting version of V3 that's good for outpainting.</li><li>V2 Stronger painterly style. High contrast and sharpness. More RPG knowledge.</li><li>V2 offset Noise Offset added making more contrast and bringing the model back to photoreal.</li><li>V2 Art Trained model. Very artsy. Strongest painterly style. Less details and bigger brush strokes to mimic digital painting style pre-AI.</li><li>V2 inpaint Inpainting version of V2 that's good for outpainting.</li><li>V1 Smoother renders with least painterly effect.</li><li>V1 inpaint Inpainting version of V1 that's good for outpainting.</li></ul> Stable Diffusion 2.1 (768) versions:<ul><li>SD 2.1 768 V1 Strong painterly style, very coherent with hands and objects. Higher native resolution and detail. Not good for nudity.</li></ul>Not as effective, but here's the LoRA if you like to use that instead: <a target="_blank" rel="ugc" href="https://civarchive.com/models/8951/a-to-zovya-rpg-artists-tools-lora">A to Zovya RPG Artist's Tools LoRA</a> Do you have requests? I've been putting in many more hours lately with this. That's my problem, not yours. But if you'd like to tip me, buy me a beer. Beer encourages me to ignore work and make AI models instead. Tip and make a request. I'll give it a shot if I can. <a target="_blank" rel="ugc" href="https://ko-fi.com/zovya">Here at Ko-Fi</a>

A-Zovya RPG Artist Tools

The Stable Diffusion 2.1 768 inpainting version. Set VAE to Automatic using 2.x models. Good for outpainting. Be sure to also download the config file and place in the same folder with the model.

aZovyaRPGArtistTools_sd21768V1Inpainting.safetensors

aZovyaRPGArtistTools_sd21768V1Inpainting.yaml

SD2.1-768 V1 Inpainting

V4 Pruned

V4 Inpainting

V4-VAE

V3+VAE

V3 Inpainting

V2 Inpaint

V1 Pruned

V2 Art

V1 inpaint

V2 offset

SD 2.1 768 V1

Warning: contains a small but significant amount of unsafe content.This model was trained by me in Dreambooth, and then merged with other models. I don’t dare to name the exact number of images in the dataset, but I guess there are more than 160000.This is a truly multi-concept checkpoint. Portraits, landscapes, 3d models, interior, objects, food, characters, animals. Thanks to a versatile approach to data captioning, the result remains predictable with short and long prompts, as well as prompts consisting of separate words and whole sentences.All examples are generated without face restoration and hires fix in the original resolution.Contains noise offset with a large value and baked <a target="_blank" rel="ugc" href="https://civarchive.com/models/87271/tone-range-compressor-vae">TRCVAE</a>, which should give more subtle control over the brightness and contrast level.警告：包含少量但重要的不安全内容。这个模型是我在 Dreambooth 中训练的，然后与其他模型合并。我不敢说数据集中的确切图片数量，但我猜有超过 160000 张。这是一个真正的多概念检查点。肖像，风景，3d 模型，室内，物体，食物，人物，动物。由于采用了多样化的数据描述方法，无论是短的还是长的提示，还是由单个单词或整个句子组成的提示，结果都保持可预测性。所有示例都是在原始分辨率下生成的，没有使用 face restoration 和 hires fix。包含 noise offset 和烘焙 <a target="_blank" rel="ugc" href="https://civarchive.com/models/87271/tone-range-compressor-vae">TRCVAE</a>，这应该能够更细致地控制亮度和对比度水平。Предупреждение: содержит малое, но существенное количество небезопасного контента.Эта модель была обучена мной в Dreambooth, а затем объединена с другими моделями. Я не решусь назвать точное число изображений в датасете, но предположу, что их более 160000.Это по-настоящему мультиконцептный чекпоинт. Портреты, пейзажи, 3d модели, интерьер, предметы, еда, персонажи, животные. Благодаря разностороннему подходу к разметке данных, результат остаётся предсказуемым и с короткими, и с длинными, и состоящими из отдельных слов, и состоящими из целых предложений промптами.Все примеры сгенерированы без face restoration и hires fix в оригинальном разрешении.Содержит noise offset с большим значением и запечённый <a rel="ugc" href="https://civarchive.com/models/87271/tone-range-compressor-vae">TRCVAE</a>, что должно давать более тонкий контроль над уровнем яркости и контраста.

Insanity (2.1 768)

insanity21768_zphyrinsanity-inpainting.ckpt

insanity21768_zphyrinsanity-inpainting.yaml

insanity21768_zphyrinsanity-inpainting.safetensors

Z_phyr Insanity (2.1 768)

ZphyrInsanity-inpainting

ZphyrInsanity

<ul><li>update - 4th Feb 2023: V15 version added</li><li>update - 19th Jan 2023: Johnson has been experimenting and passed this to me to include on this listing: djzGingerTomCatV21-512-inpainting please use an SD-V21-512-inpainting.yaml config (included below) we recommend using latent nothing mask mode. </li></ul>This Model combines two GingerTomCat datasets, created by DriftJohnson. It is then reinforced by two textual inversions; the <a target="_blank" rel="ugc" href="https://civarchive.com/models/1917/kittypic-for-sd-2x">KittyPics </a>embedding by <a target="_blank" rel="ugc" href="https://civarchive.com/user/stille_willem">stille_willem </a>&amp; the <a target="_blank" rel="ugc" href="https://civarchive.com/models/4044/doctor-diffusions-point-e-negative-embedding">Point-E</a> negative embedding by <a target="_blank" rel="ugc" href="https://civarchive.com/user/doctor_diffusion">Doctor_Diffusion</a>3x3 grid used both embeddings, 2x2 grid shows same seed outputs with some suggested upscaler settings.<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/caf909d4-2a39-4604-22c7-1b08a0cab100/width=525" /><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9d9a407c-41a8-4da0-0cd2-72d18dc1a400/width=525" />showcase credit: <a target="_blank" rel="ugc" href="https://civarchive.com/user/DriftJohnson">DriftJohnson</a> "strong style" Models are intended to be merged with each other and any model for Stable Diffusion 2.1 -- although you can also use these without the Trigger word like any modelI recommend merging with 0.5 (50/50 blend) then using prompt weighting to control the Aesthetic gradient.example merged model prompt with automatic1111:(GingerTomCat:1.2) (yourmodeltoken:0.8)if you drop the "djz" and the "V21" what remains is the token you need to call up the concept in the model. All examples shown were the Raw Token, no other words. Tokens are case sensitive and in almost all models it will match the filename.It is possible to merge these models with each other using a different value. It is possible to pair models and then merge those resulting models. In this way we can blend abstract concepts together and then weight the tokens to achieve the result we may wish to create. Of course to eliminate all those tokens, you can simply train a new custom model from the outputs, which means you are back to a single token.A video explanation will follow, but for now the above explanation should do. We are focused on getting as many style/aesthetic models into artist hands to enhance the creativity already at their finger tips.Art Freedom for all!![all original artwork used for training with <a target="_blank" rel="ugc" href="https://twitter.com/MushroomFleet/status/1609633158864961539">full permission from Drift Johnson</a>]

djz GingerTomCat V21-768 / V21-512-inpainting / V15

please use the inpainting config for SD V2-1-512-inpainting, it is included here, ensure the filenames are matching after you download :)

djzGingertomcatV21768V21_21512-inpainting.yaml

djzGingertomcatV21768V21_21512-inpainting.safetensors

2-1-512-inpainting

2-1-768

<ul><li>31st January 2023 Update: djzAirlockV15 released!</li><li>19th January 2023 Update: djzAirlockV21-512-inpainting released! you will need the 2-1-512-inpainting.yaml config, so grab that too!</li></ul>2.1-768 model: Example merging <a target="_blank" rel="ugc" href="https://civarchive.com/models/3410/rick-roll-style">Rick-Roll</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/4007/djz-junglepunk-v21">JunglePunk</a> &amp; <a target="_blank" rel="ugc" href="https://civarchive.com/models/4043/djz-airlock-v21">Airlock</a>:<img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ce27e991-6393-411b-29da-cffddc77d400/width=525" />credit: <a target="_blank" rel="ugc" href="https://civarchive.com/user/RickInversion">RickInversion</a> These "strong style" Models are intended to be merged with each other and any model for Stable Diffusion 2.1I recommend merging with 0.5 (50/50 blend) then using prompt weighting to control the Aesthetic gradient.example merged model prompt with automatic1111:(Airlock:1.2) (yourmodeltoken:0.8)if you drop the "djz" and the "V21" what remains is the token you need to call up the concept in the model. All examples shown were the Raw Token, no other words. Tokens are case sensitive and in almost all models it will match the filename.It is possible to merge these models with each other using a different value. It is possible to pair models and then merge those resulting models. In this way we can blend abstract concepts together and then weight the tokens to achieve the result we may wish to create. Of course to eliminate all those tokens, you can simply train a new custom model from the outputs, which means you are back to a single token.A video explanation will follow, but for now the above explanation should do. We are focused on getting as many style/aesthetic models into artist hands to enhance the creativity already at their finger tips.Art Freedom for all!![all original artwork used for training with <a target="_blank" rel="ugc" href="https://twitter.com/MushroomFleet/status/1609633158864961539">full permission from Drift Johnson</a>]

djz Airlock V21-768, V21-512-inpainting, V15

ensure you take the correct config! demo photos were random pictures i took with a camera. you can do better easily!

djzAirlockV21768V21512_21512-inpainting.yaml

djzAirlockV21768V21512_21512-inpainting.safetensors

Files with SHA256: 821a51ee6a8f3a62e7f2002557fcb4f4ae97a2ab3cb3fb85942897d536fa8b2c