Update README.md
Browse files
README.md
CHANGED
@@ -53,6 +53,13 @@ As with most other AI image generative models, the flux-dev model is biased towa
|
|
53 |
|
54 |
The goal with these boring reality trained loras is to not only bring out better photorealistic images but to push the model to show how much knowledge and information it can actually place in a single generated image.
|
55 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
56 |
**Primary Goals for Boreal-FD**
|
57 |
- Reduce how often shallow depths of field appear in images
|
58 |
- More dynamic poses
|
|
|
53 |
|
54 |
The goal with these boring reality trained loras is to not only bring out better photorealistic images but to push the model to show how much knowledge and information it can actually place in a single generated image.
|
55 |
|
56 |
+
**Update 08/21**
|
57 |
+
I am still exploring new ways to train this model/dataset. For the timebeing, the faded dot issue remains in these LoRAs as these older models have performed a better job of learning the concept than any of the subsequent runs I have done.
|
58 |
+
The new Schnell version may be released before updating this Dev model.
|
59 |
+
As there are multiple safetensor files, you may need to resolve which model to use if you are using the hf_lora string with something like: **kudzueye/Boreal/resolve/main/boreal-flux-dev-lora-v04_1000_steps.safetensors**.
|
60 |
+
I will try to seperate models going forward, though it is not the best strategy when you need to manually choose between undertrained/overtrained LoRAs on top of their strength for each image.
|
61 |
+
|
62 |
+
|
63 |
**Primary Goals for Boreal-FD**
|
64 |
- Reduce how often shallow depths of field appear in images
|
65 |
- More dynamic poses
|