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

Style of Moebius 🌊 Flux.1-D

90
655
109
60
Updated: Feb 19, 2025
stylegraphic noveljean giraudmoebiusflux.1flux1.d
Verified:
SafeTensor
Type
LoRA
Stats
655
109
Reviews
(90)
Published
Aug 18, 2024
Base Model
Flux.1 D
Training
Steps: 6,000
Usage Tips
Strength: 1
Trigger Words
In style of Moebius
Hash
AutoV2
6109297D51
Uploads - 19
19
Downloads - 4834
4.8k
seruva19's Avatar
seruva19
Joined Jan 3, 2023
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.
Please support civitai and creators by disabling adblock
ComfyUI_temp_jpsna_00067_.png
ComfyUI_temp_jpsna_00086_.png
ComfyUI_temp_jpsna_00118_.png
ComfyUI_temp_jpsna_00142_.png
ComfyUI_temp_jpsna_00144_.png
ComfyUI_temp_jpsna_00161_.png
ComfyUI_temp_jpsna_00165_.png
ComfyUI_temp_jpsna_00209_.png
ComfyUI_temp_jpsna_00223_.png
ComfyUI_temp_jpsna_00232_.png
ComfyUI_temp_jpsna_00237_.png
ComfyUI_temp_jpsna_00258_.png
ComfyUI_temp_jpsna_00259_.png
ComfyUI_temp_jpsna_00281_.png
ComfyUI_temp_jpsna_00285_.png
ComfyUI_temp_jpsna_00307_.png
ComfyUI_temp_jpsna_00318_.png
ComfyUI_temp_jpsna_00332_.png
Images hidden due to mature content settings

Overview

This F1 LoRA aims to reproduce the distinctive style of Jean Giraud, a famous French artist who made significant contributions to the Franco-Belgian bandes dessinées (BD) tradition and is better known under the pseudonym Moebius. His unique style, characterized by intricate line work, expansive worlds and visionary, surreal and imaginative designs left an indelible mark on the genre.

Usage

All showcase images were generated with the following settings, which are recommended:

Prompts: started with "In style of Moebius."
Model: flux1-dev (fp8e4m3fn)
Text Encoder: t5pxxl_fp16
Sampler: euler
Scheduler: 24 steps (normal)
Flux Guidance: 4
LoRA Strength: 1

Training

LoRA was created using the ai-toolkit on an RTX 3090. The fine-tuning process utilized 362 images of varying quality, all autocaptioned with Florence-2 ("<MORE_DETAILED_CAPTION>" task), with addition of "In style of Moebius. " into the beginning of each caption. Training hyperparameters were:

Rank/alpha: 32/32
Batch Size: 1
Steps: 6000
Learning Rate: 1e-4
Optimizer: adamw8bit

Discussion

SD
sdf34gdssdg4

First time I see an AI getting close to Moebius style.