Sign In
home
models
images
videos
posts
articles
bounties
tools
challenges
events
shop

Westeros Continent - Game of Thrones / House of Dragons

27
288
338
9
Updated: Aug 31, 2024
stylemedievalgame of thronesfantasydragonwesteros
Verified:
SafeTensor
Type
LoRA
Stats
224
338
Reviews
(17)
Published
Aug 31, 2024
Base Model
Flux.1 D
Training
Steps: 5,000
Trigger Words
Westeros
Game of thrones
dragon
Hash
AutoV2
E97049AAF4
default creator card background decoration
DirectorKobayashi's Avatar
DirectorKobayashi
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

V2.0 UPDATE:
Pumped up training with 40 extra images, training resolution from 1024px to 1280px, network size from 16 to 64, 5000 steps.
I think the quality of the output has improved drastically!!!

V1.0:
I trained this lora for Flux locally in ComfyUI using a dataset of 100 images, all verbally described (thanks to Joy caption pre alpha) and trained for 2480 steps.

In the prompt you have to describe as many details as you can.

Use the keywords Westeros, Game of Thrones, Fantasy to activate the Lora.

About settings:


Fantastic creatures and locations work best with low cfg (2 to 3) - if the shot has some human characters it's better with mid-high cfg (3.5 to 7).
I think a lora weight of 0.8-0.85 gives good results.

You can find my workflow in the image without any watermarks or post-production.
Enjoy and please share here your creations!!!