Update README.md

README.md

@@ -21,6 +21,10 @@ library_name: diffusers
 
 # AstolfoMix (Baseline) #
 
+- A (baseline) merge model focusing on [absurdres](https://www.urbandictionary.com/define.php?term=absurdres), *and let me wait for a big anime SDXL finetune.*
+- Behind the "absurdres", the model should be very robust and capable for most LoRAs / embeddings / addons you can imagine.
+- The image below is 2688x1536 without upscaler. With upscaler, it reaches 8K already.
+
 ![231077-3179120050-2688-1536-4.5-64-20230917093912.jpg](https://huggingface.co/6DammK9/AstolfoMix/resolve/main/231077-3179120050-2688-1536-4.5-64-20230917093912.jpg)
 
 ```
@@ -30,16 +34,22 @@ Negative prompt: (worst:0), (low:0), (bad:0), (exceptional:0), (masterpiece:0),
 Steps: 64, Sampler: Euler, CFG scale: 4.5, Seed: 3179120050, Size: 1344x768, Model hash: f52ee1e6b3, Model: vcbp_mtd8_cwl-sd, VAE hash: 551eac7037, VAE: vae-ft-mse-840000-ema-pruned.ckpt, Denoising strength: 0.7, Clip skip: 2, Hires upscale: 2, Hires steps: 32, Hires upscaler: Latent, Dynamic thresholding enabled: True, Mimic scale: 1, Separate Feature Channels: False, Scaling Startpoint: MEAN, Variability Measure: AD, Interpolate Phi: 0.3, Threshold percentile: 100, Version: v1.6.0
 ```
 
-- Probably the best "baseline model" I've ever seen.
-- Baseline Model: `vcbp_mtd8_cwl-sd.safetensors`
+- Current version: `vcbp_mtd8_cwl-sd.safetensors` (merge of 5 models)
 - [Receipe Models: Merging UNETs into SD V1.4](https://huggingface.co/6DammK9/bpmodel-sd14-merge)
 - ["Roadmap" / "Theory" in my Github.](https://github.com/6DammK9/nai-anime-pure-negative-prompt/blob/main/ch05/README.MD)
-- Receipe: **Uniform merge** of 4 models. M=0.5 (`vbp23-cbp2-sd-v1.4`) then M=0.33 (`vcbp_mzpikas_tmnd-sd-v1.4`) then M=0.25 (`vcbp_mzpt_d8-sd-v1.4`)
 - Recommended prompt: "SD 1.4's Text Encoder"
 - Recommended resolution: **1024x1024 (native T2I), HiRes 1.75x (RTX 2080Ti 11GB)**
 - It can generate images **up to 1280x1280 with HiRes 2.0x ([Tesla M40 24GB](https://github.com/6DammK9/nai-anime-pure-negative-prompt/blob/main/ch04/readme.md#chapter-04b-extra-making-use-of-m40-24gb-for-generating-large-images))**, but the yield will be very low and time consuming to generate a nice image.
 - Recommended CFG: 4.5 (also tested on all base models), 6.0 (1280 mode)
 
+## Receipe ##
+
+- **Uniform merge.** M = 1 / "number of models in total".
+- M=0.5 (`vbp23-cbp2-sd-v1.4`)
+- M=0.33 (`vcbp_mzpikas_tmnd-sd-v1.4`)
+- M=0.25 (`vcbp_mzpt_d8-sd-v1.4`)
+- M=0.2 (`vcbp_mtd8_cwl-sd`)
+
 ![xyz_grid-0083-342433587-10240-1656-4.5-48-20230916031640.jpg](https://huggingface.co/6DammK9/AstolfoMix/resolve/main/xyz_grid-0083-342433587-10240-1656-4.5-48-20230916031640.jpg)
 
 ## License ##