Deep Dark Pony

I’ve finally jumped on the Pony Train. Kind of. Yes this is technical a v6 Pony model. And yes, this is also technically an SDXL 1.0 model. It’s a hybrid. It prompts and follows prompts like SDXL but has the understanding and capabilities of a very naughty pony model. There is no need for extra quality keywords, no scores to list, and no need to use a style or explicit prompts. It just works.

Prompt with Danbooru tags. Prompt with natural language. Do a combination. Use a pony prompt. Use prompts from your favorite model. Use a haiku or a dirty limerick. It doesn’t matter. It’ll do any NSFW you could desire. It’ll do SFW. It’ll take your childhood memories and ruin them. Don’t prompt it if you don’t want to see it.

The default style is meant to replicate Deep Dark Hentai Mix, so you’ll see an art style that ranges from well drawn anime to semi-realistic, often with backgrounds that are more real than drawn. In my opinion, its not quite there, but stylistically, its much closer to DDHM than Pony. Pony also models tend to exaggerate anatomy, but in Deep Dark Pony it only really comes out when prompting for it. 

Usage Tips and Prompting:

• Most versions of the model are very picky about sampler settings. I typically use any of the following settings, all of which work great: DPM++ SDE Karras, DPM++ 2M Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, Euler a, DPM adaptive. Other samplers may work, but if you are getting images that don’t look fully diffused, that’s the issue.

• It's part Turbo model, so Steps and CFG will scale. On the low end, use 6 to 10 Steps and 1.5 to 3.5 CFG Scale. If you like higher steps, you can go up to 30 steps and a 5 CFG scale. All showcase images are 8 steps and 2.5 CFG scale. Clip skip doesn’t matter.

• LoRAs: So far Pony LoRAs work and SDXL LoRAs work. Realistic LoRAs seem to work as well as Anime. I’ve only experimented with 10 or so, so I can’t promise you they all will work, but so far, so good.

• Prompt any way and any style you want. So far, any prompt I’ve seen listed for any model I normally use will work. No need for Pony’s scores. If you want to improve the quality and resolution of images, use tags like high resolution, highres, best quality, 4k. Intricately detailed and amazing details work to add extra complexity. Cinematic pushes images to 3D and makes some generations realistic. Don't use masterpiece, immaculate, absurdres, absurd res, 85mm -- they all do more harm than good. No negatives are needed as a standard prompt; things like low quality or bad anatomy aren't necessary. 

• Versions 4.0 and 5.0 of DDHM had almost eliminated watermarks, especially with select keywords in the prompting. However, when mixing with Pony, I noticed that new watermarks in the upper corners sometimes appear. For me it happens infrequently, but I bet its related to a keyword in the prompt (either decreasing them for me or increasing them for you). Weird that Pony made it worse. I hope for better luck in the next version.

• Some of the models included in thie merge are capable of producing some really beautiful artwork. Even if you aren’t into hentai, much of the time you will be able to generate some really great SFW images with very good anatomical details. 

• This model is not intended to produce illegal acts and images. However, to be 100% safe, add "children, loli, child, kids, teens" to your negative prompt. Pony can be unpredictable, so safe is better than sorry.

Model Credits

The following models have been or are currently used in a version of Deep Dark Pony. Credit goes to the model creators (and of course, the creators of the models that were used in their merges/training as well).

• Deep Dark Hentai Mix - NSFW Anime – https://civarchive.com/models/221751

• Pony Diffusion V6 XL – https://civarchive.com/models/257749?modelVersionId=290640

Legal Stuff

AI model output may sometimes be unpredictable and is not guaranteed to match the input prompt nor to be of any given nature or quality. Do not attempt to generate content that is illegal in your legal jurisdiction or within your country or district of residence. By using this model, you agree that the creator of this model shall in no event be responsible or liable to you or to any third party, whether under contract, warranty, tort (including negligence), strict liability, indemnity or other theory, for any direct, indirect, special, incidental, consequential, exemplary, liquidated or punitive damages or any other damages, including but not limited to loss of profit, revenue or business, cost of substitute procurement, arising in whole or in part from your use of (or inability to use) the model.

Model Restrictions:

To be clear, this model is a derivative of the SDXL 1.0 model and the following restrictions are part of the 2023 Stability AI CreativeML Open RAIL++-M License dated July 26, 2023, with applies to this model and all models that use any component or derivatives of this model. You agree not to use the Model or Derivatives of the Model:

• In any way that violates any applicable national, federal, state, local or international law or regulation; 

• For the purpose of exploiting, harming or attempting to exploit or harm minors in any way; 

• To generate or disseminate verifiably false information and/or content with the purpose of harming others; 

• To generate or disseminate personal identifiable information that can be used to harm an individual; 

• To defame, disparage or otherwise harass others; 

• For fully automated decision making that adversely impacts an individual’s legal rights or otherwise creates or modifies a binding, enforceable obligation; 

• For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics; 

• To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm; For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories; 

• To provide medical advice and medical results interpretation; 

• To generate or disseminate information for the purpose to be used for administration of justice, law enforcement, immigration or asylum processes, such as predicting an individual will commit fraud/crime commitment (e.g. by text profiling, drawing causal relationships between assertions made in documents, indiscriminate and arbitrarily-targeted use).

Clarifications on the permissible use of this model:

• This model is a merge. Therefore it inherits the permissions and restrictions of all models used in its creation.

• You may not make merges, extractions, or derivatives of this model and use a less restrictive license than the 2023 Stability AI CreativeML Open RAIL++-M License.

• This model may be run and hosted for download on Civitai.com. At this time, no other organization or individual is authorized to host or distribute this model.

• You may not sell this model or any of this model’s merges or derivatives.

• You must credit this model’s creator if you use this model to create a new merged, mixed, trained, or extracted model.

• You may sell images you generate with this model.

• You may share merges using this model as long as any merge has the same or more restrictive permissions.