Update README.md

README.md

@@ -19,15 +19,17 @@ metrics:
 - character
 ---
 
-<img src="https://huggingface.co/hollowstrawberry/hololive/resolve/main/out.jpg" width="600"/>
+<img src="https://huggingface.co/hollowstrawberry/hololive/resolve/main/out.jpg" width="800"/>
 
 
 # Models
 
+These were mixed with a model trained on tens of thousands of vtuber images. The 3.1 version knows more non-hololive youtubers, but may have lower quality.
+
 * [HoloCreamsicle](https://huggingface.co/grugger/chubas/blob/main/models/hll1/HoloCreamsicle.pruned.safetensors) ([aom2_hard](https://huggingface.co/WarriorMama777/OrangeMixs/blob/main/Models/AbyssOrangeMix2/AbyssOrangeMix2_hard.safetensors) + hll1)
 * [MightyMix](https://huggingface.co/grugger/chubas/blob/main/models/hll1/mighty_mix.pruned.safetensors) (HoloCreamsicle + [7th_anime_v2_A](https://huggingface.co/syaimu/7th_Layer/blob/main/7th_anime_v2/7th_anime_v2_A.ckpt))
 * [Qcha-hllv1](https://huggingface.co/Aotsuyu/Qcha/blob/main/Qcha-hllv1.safetensors) (many merges, heavy anime style)
-* [A_2v_eciuJht9](https://huggingface.co/grugger/chubas/blob/main/models/A_2v_eciuJht9.pruned.safetensors) (aka MightyMix_hll3.1, more vtubers)
+* [A_2v_eciuJht9](https://huggingface.co/grugger/chubas/blob/main/models/A_2v_eciuJht9.pruned.safetensors) (aka MightyMix_hll3.1)
 * Manually mixing [hll1](https://huggingface.co/grugger/chubas/blob/main/models/mirrors/hll-epoch-000015.safetensors)/[hll3.1](https://huggingface.co/grugger/chubas/blob/main/models/mirrors/hll3vtubers-last-pruned.safetensors) with your desired model
 
 <details>
@@ -38,13 +40,36 @@ metrics:
 
 &nbsp;
 
-# All Vtubers
+# Loras (and TIs)
+
+All known holo Loras will be linked to and backed up here. Put them in your `stable-diffusion-webui/models/Lora` folder. Use them in your prompt like this: `<lora:filename:1>`
+
+Most Loras work best with a weight of `0.7` as the default of 1 might fry your image.
+
+## All Vtubers
+
+You can use it with any model, but may have lower quality than Loras for individual vtubers or a full merged model.
 
 * [hll1 and hll3.1 extracted into a Lora](https://civitai.com/models/10664/all-vtubers-lora)
+* [Original source](https://mega.nz/folder/rBVGjIpQ#WCl9DL65mMRyaBmyIPA9WQ)
 
 &nbsp;
 
-# Outfits (LORA/TI)
+## Small Loras
+
+These were resized to a quarter of their original size. Give them a higher weight than their full counterparts to preserve detail. Find them in this repo.
+
+<details>
+<summary>Click here for a comparison</summary>
+<br>
+<img src="https://huggingface.co/hollowstrawberry/hololive/resolve/main/lora_small.jpg" width="800"/>
+</details>
+
+&nbsp;
+
+## Outfits
+
+Main purpose of this repo.
 
 ### HoloEN
 
@@ -95,7 +120,7 @@ metrics:
 * Usada Pekora [(Default)](https://civitai.com/models/9567/usada-pekora) [(2)](https://discord.com/invite/touhouai) [(Seifuku, Unreleased Bunnysuit)](https://mega.nz/folder/y3BnWajK#PxqXj4wXbRS8or7j63068g/folder/zroGgbxA)
 * Shiranui Flare
 * Shirogane Noel
-* Houshou Marine (Unknown)
+* Houshou Marine ({TI} Unknown)
 * Uruha Rushia [(Default, Black, Pink, White, Neko, Bikini)](https://huggingface.co/Lineins/All_Uruha_Rushia_LoRA) [({TI} Default, Black, Pink, Neko)](https://huggingface.co/soknife/embeddings)
 * Tsunomaki Watame [(Default)](https://civitai.com/models/10017/tsunomaki-watame-tokoyami-towa) [(2)](https://civitai.com/models/9430/tsunomaki-watame-or-hololive)
 * Tokoyami Towa [(Default)](https://civitai.com/models/10017/tsunomaki-watame-tokoyami-towa) [(2)](https://mega.nz/folder/OppA0QDI#w0_KOY0CyTpPtG6xopKwOw/folder/LwgDjAjK)