Update README.md

README.md

@@ -40,8 +40,24 @@ tags:
 - gguf-connector
 ---
 # **gguf quantized version of krea**
-- run it with diffusers (see example inference below)
+- run it straight with `gguf-connector`
+- opt a `gguf` file in the current directory to interact with by:
+```
+ggc k
+```
+>
+>GGUF file(s) available. Select which one to use:
+>
+>1. flux1-krea-dev-q2_k.gguf
+>2. flux1-krea-dev-q4_0.gguf
+>3. flux1-krea-dev-q8_0.gguf
+>
+>Enter your choice (1 to 3): _
+>>>
+
+![screenshot](https://raw.githubusercontent.com/calcuis/gguf-pack/master/k4.png)
 
+- run it with diffusers (see example inference below)
 ```py
 import torch
 from transformers import T5EncoderModel
@@ -89,40 +105,6 @@ image.save("output.png")
 
 ![screenshot](https://raw.githubusercontent.com/calcuis/comfy/master/krea.png)
 
-## **run it with gguf-connector**
-- able to opt - 2-bit, 4-bit or 8-bit model for dequant while executing:
-```
-ggc k4
-```
-
-![screenshot](https://raw.githubusercontent.com/calcuis/gguf-pack/master/k4.png)
-
-### 🐷 Krea 7 Image Generator (connector mode)
-- opt a `gguf` file straight in the current directory to interact with
-```
-ggc k7
-```
-
-> device auto-detection logic applied (different by system)
->>>
->
->GGUF file(s) available. Select which one to use:
->
->1. flux1-krea-dev-q2_k.gguf
->2. flux1-krea-dev-q4_0.gguf
->3. flux1-krea-dev-q8_0.gguf
->
->Enter your choice (1 to 3): _
->>>
-
-- k7 supports all t, k, s quants plus iq4_nl, iq4_xs, iq3_s, iq3_xxs and iq2_s
-- compatible with kontext and flux1-dev files as well; try it out!
-
-### All-in-one flux connector
-```
-ggc k
-```
-
 ### **reference**
 - base model from [black-forest-labs](https://huggingface.co/black-forest-labs)
 - diffusers from [huggingface](https://github.com/huggingface/diffusers)