<h2>Final Fantasy VII Remake style (UE4) LoRA</h2>Making models can be expensive. Do you like what I do? Consider supporting me on <a target="_blank" rel="ugc" href="https://www.patreon.com/Lykon275">Patreon</a> 🅿️ or feel free to <a target="_blank" rel="ugc" href="https://snipfeed.co/lykon">buy me a coffee</a> ☕ UPDATE: Retrained on NED.<ul><li>For v1 a weight around 0.65 works fine from my tests. Triggers with <code>ff7r style</code>.</li><li>For NED version use weight 1 on <a target="_blank" rel="ugc" href="https://civarchive.com/models/10028/neverending-dream-ned?modelVersionId=11925">Never Ending Dream (v1.0)</a>. Triggers with <code>ff7r style</code>. I'll maybe train another one on NED 1.22 to make it accurate on that one too. </li></ul>Also this came up kind of biased towards FF7 characters, so while this is mainly a style LoRA, it could replicate characters from the main cast of the game fairly well without embeddings or additional LoRA.V1 original description: This is kind of experimental. I wanted to see if I could finetune an anime model to replicate the style of a game in Unreal Engine. I think this was kind of a success. It works on AbyssOrangeMixV2 with a normal negative prompt, or on AnyLoRA with my usual negative prompt with negative embeddings (but don't mix them up, as that neg prompt doesn't work well on AOM with this LoRA). See the examples for the various comparisons. How to use LoRA's in auto1111:<ul><li>Update webui (use <code>git pull</code> <a target="_blank" rel="ugc" href="https://www.youtube.com/embed/mn8fMF10XN4?start=31&amp;end=60">like here</a> or redownload it)</li><li>Copy the file to <code>stable-diffusion-webui/models/lora</code></li><li>Select your LoRA like in <a target="_blank" rel="ugc" href="https://www.youtube.com/watch?v=-bMeyXOZwN0">this video</a></li><li>Make sure to change the weight (by default it's <code>:1</code> which is usually too high)</li></ul>

Final Fantasy VII Remake Style (Unreal Engine 4) LoRA

Trained on NED. Does the accurate style when used on NED, a more anime style when used on anime models. Like a UE4 to anime conversion.

ff7r_style_ned_offset.safetensors

NED offset

UPDATE: Retrained on NED.<ul><li>For v1 a weight around 0.65 works fine from my tests. Triggers with ff7r style.</li><li>For NED version use weight 1 on Never Ending Dream (v1.0). Triggers with ff7r style. I'll maybe train another one on NED 1.22 to make it accurate on that one too.</li></ul>Also this came up kind of biased towards FF7 characters, so while this is mainly a style LoRA, it could replicate characters from the main cast of the game fairly well without embeddings or additional LoRA.V1 original description: This is kind of experimental. I wanted to see if I could finetune an anime model to replicate the style of a game in Unreal Engine. I think this was kind of a success. It works on AbyssOrangeMixV2 with a normal negative prompt, or on AnyLoRA with my usual negative prompt with negative embeddings (but don't mix them up, as that neg prompt doesn't work well on AOM with this LoRA). See the examples for the various comparisons.

Final Fantasy VII Remake Style (Unreal Engine 4) L

For v1 a weight around 0.65 works fine from my tests. Triggers with ff7r style.

For NED version use weight 1 on Never Ending Dream (v1.0). Triggers with ff7r style. I'll maybe train another one on NED 1.22 to make it accurate on that one too.

Also this came up kind of biased towards FF7 characters, so while this is mainly a style LoRA, it could replicate characters from the main cast of the game fairly well without embeddings or additional LoRA.

V1 original description:
This is kind of experimental. I wanted to see if I could finetune an anime model to replicate the style of a game in Unreal Engine. I think this was kind of a success. It works on AbyssOrangeMixV2 with a normal negative prompt, or on AnyLoRA with my usual negative prompt with negative embeddings (but don't mix them up, as that neg prompt doesn't work well on AOM with this LoRA). See the examples for the various comparisons.


How to use LoRA's in auto1111:

Update webui (use git pull like here or redownload it)

Copy the file to stable-diffusion-webui/models/lora

Select your LoRA like in this video

Make sure to change the weight (by default it's :1 which is usually too high)

Files with SHA256: 36561d745eff71bda45d968afbcd6f80dde77ea0be6738e5c2f6ecdb65517540