Note: The sample images are not representative of the specific style, as these models should adapt to various styles based on the checkpoint and other Loras used. The goal of these models is to capture the various male body types.<h1 id="male-mix-pony-edition">Male Mix Pony Edition</h1>For those who don't know, Pony already does a decent job capturing most male body types, so in some cases a lora like this isn't even necessary. This began as an SDXL lora, as SDXL doesn't seem to have as much training on male bodies, and was just converted to Pony as a test to add a little more variety to what the base model already had. The main issue with Pony is that it seems like there was a very limited "photo/realistic" image set in the training data, so it does tend to give a similar face to most males done in a realistic style from my testing.v1 and v2 aren't much different in overall functgion, I just got some stylistic changes that I thought were interesting. v3 is a merge of v1&amp;2 with some additional style loras, so it has a much different style, but doesn't do photo prompts as well as the other two.NOTE: Unfortunately the merge for v3 didn't work correctly, so it doesn't play well with some other older loras.v4 uses a more diverse data set, that also includes photo/realistic images. Because of this I feel like I was able to capture the different body types better (finally some differentiation between femboy and androgynous), and also realistic images look much better (though it still leans toward whatever realistic male face the base pony model was trained on, it does try to break away from that more).v5 uses a completely new data set, which also contains photo/realistic images like v4. This data set is smaller overall, but has a more consistent style to it, based on the Illustrious checkpoint merge I made, Zabadu's Mix. However, due to it's smaller size it still has some of the body type overlap from the base model. I also added in the 3 body type captions for 'Bara', 'Plump' and 'Chubby'.v1<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ef43ab81-b024-490b-8ae9-1758ecc9f85c/width=525/ef43ab81-b024-490b-8ae9-1758ecc9f85c.jpeg" />v2<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7f21c4a7-6d9a-4359-9251-bae2cc9ad5ea/width=525/7f21c4a7-6d9a-4359-9251-bae2cc9ad5ea.jpeg" />v3<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5d6c1e16-33e9-4391-8488-d0c9d1eef8f0/width=525/5d6c1e16-33e9-4391-8488-d0c9d1eef8f0.jpeg" />v4<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/47b3b7ff-c7ea-43db-a209-d157e432a6aa/width=525/47b3b7ff-c7ea-43db-a209-d157e432a6aa.jpeg" />v5<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0efe7b50-599c-4e15-a41a-26b98f5a0482/width=525/0efe7b50-599c-4e15-a41a-26b98f5a0482.jpeg" />+'Bara', 'Plump', and 'Chubby'<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/bf82b805-5d82-4f8f-8cd5-daf6e263011a/width=525/bf82b805-5d82-4f8f-8cd5-daf6e263011a.jpeg" />The 'Femboy' and 'Androgynous' tags will get you similar results, but using one instead of the other can at least give you a different output. The same goes for the 'Toned' and 'Muscular' tags. And the 'Twink' and 'Skinny' tags have the some overlap.For v5 the three additional body type tags, 'Bara' gives similar results to 'Muscular', but generally with larger muscles, and 'Plump' and 'Chubby' give vary similar results to each other, with 'Chubby' being slightly larger.v4 and v5 have the best overall differentiation between the different body type tags.The training data for v1-3 didn't include any realistic images or photos, so without using a checkpoint, or additional lora specifically made for photos, you likely will only get the digital art style. v4 and v5 did include realistic images, but this still works best with a checkpoint that can handle realism.v1<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b32d8055-7bd0-4096-9d90-b0f811285385/width=525/b32d8055-7bd0-4096-9d90-b0f811285385.jpeg" />v2<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e7037cf5-0d4c-42da-9dd6-378c969083bb/width=525/e7037cf5-0d4c-42da-9dd6-378c969083bb.jpeg" />v3<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a09821b6-3651-43ac-8d8d-a4a72568789f/width=525/a09821b6-3651-43ac-8d8d-a4a72568789f.jpeg" />v4 - As stated, I did add photo/realistic images to the data set, so this model does much better with realism.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/30b0849e-ff1a-4208-8aaf-b200fcc0dc4b/width=525/30b0849e-ff1a-4208-8aaf-b200fcc0dc4b.jpeg" />v5 - Also had photo/realistic images in the data set, but still works better with artistic styles.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/98a220e3-d4e9-41e5-a9f8-fba247727613/width=525/98a220e3-d4e9-41e5-a9f8-fba247727613.jpeg" />+'Bara', 'Plump', and 'Chubby'<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/fd866ded-f454-4d3b-a117-00386a368d5f/width=525/fd866ded-f454-4d3b-a117-00386a368d5f.jpeg" /><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">These models were trained using the Pony Diffusion V6 XL <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">checkpoint, and all art style generations were made using the same checkpoint, while all the photorealistic generations used ONE FOR ALL «Pony Fantasy» DPO+VAE.Images for v1-4 were made using Automatic1111, images for v5 were made using Fooocus.All of my models may be used for other training and merges.

Male Mix (Pony)

Larger dataset, better captioning, more accurate body types.

MaleMixPDv4.safetensors

Files with SHA256: 77570cfe9a4ea2bcd902af65fd8b76321aa7a3a393a8171ec0aae485f06803d8