The Fux is upon us! Civit training has blessed us with the powers to unlock the Fux.Any ⚡ buzz ⚡ donated to the model goes right back into training. 🙏<h3 id="use-v0.2-it's-highly-improved-nsfw-compared-to-v0.1.-eoey3nq2q">Use v0.2, it's highly improved NSFW compared to v0.1.</h3>This is a work in progress NSFW enabling LoRA.The goal is to unlock as much NSFW capacity as possible inside a LoRA.It's early versions still, but it is already improving the checkpoints quite a lot.Recommended checkpoint: <a target="_blank" rel="ugc" href="https://civarchive.com/models/645943">FluxUnchained</a>.The model is trained on synthetic data only. No actual photos were used.All images in the model preview are using the LoRA.To compare with/without the LoRA, compare the 2 pinned posts below.The same prompt/seed were used, the only difference is using / not using the LoRA.<a target="_blank" rel="ugc" href="https://civarchive.com/posts/5510085">Post 1</a>. <a target="_blank" rel="ugc" href="https://civarchive.com/posts/5510508">Post 2</a>. You may want to open them in new tabs to compare.Feedback and tips (both kinds), are welcomed and appreciated.

FuX Capacity 0.2 [Realistic] [NSFW Flux LoRA]

v0.2Greatly enhanced NSFW capabilities.A lot of style is lost and it goes towards realism.The model still leads more towards realism and can add some depth to anime images and has poor ethnicity recognition. This will be improved in future versions.

FuxCapacity0.2.safetensors

v0.2

v0.1

The Fux is upon us! Civit training has blessed us with the powers to unlock the Fux.Any ⚡ buzz ⚡ donated to the model goes right back into training. 🙏<h3 id="use-v0.2-it's-highly-improved-nsfw-compared-to-v0.1.-eoey3nq2q">Use v0.2, it's highly improved NSFW compared to v0.1.</h3>This is a work in progress NSFW enabling LoRA.The goal is to unlock as much NSFW capacity as possible inside a LoRA.It's early versions still, but it is already improving the checkpoints quite a lot.Recommended checkpoint: <a target="_blank" rel="ugc" href="https://civitai.com/models/645943">FluxUnchained</a>.The model is trained on synthetic data only. No actual photos were used.All images in the model preview are using the LoRA.To compare with/without the LoRA, compare the 2 pinned posts below.The same prompt/seed were used, the only difference is using / not using the LoRA.<a target="_blank" rel="ugc" href="https://civitai.com/posts/5510085">Post 1</a>. <a target="_blank" rel="ugc" href="https://civitai.com/posts/5510508">Post 2</a>. You may want to open them in new tabs to compare.Feedback and tips (both kinds), are welcomed and appreciated.

Files with SHA256: 3fe58d017200748dc4cb674c3e02d15fc1935f79307cb184680768faf6d4937b