Download
1 variant available
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.
π Midjourney Illustration Style for Flux π
β¨This Midjourney-style LoRA for Flux is inspired by the signature aesthetic of Midjourney β dreamy, surreal, richly layered with dramatic lighting and fine-grained detail. It is perfect for creators looking to bring that bold, cinematic visual language into the Flux ecosystem.
π₯ Features & Ideal Use Cases
β
Fine Style Reproduction: Captures the dreamlike atmosphere and crisp detail typical of Midjourney, resulting in rich compositions and artistic expressiveness
β
Wide Style Versatility: Works well with fantasy, sci-fi, mechanical, classical, cyberpunk, and many other visual genres
β
Strong Visual Impact: Enhances lighting contrast, fog, volumetric effects, and structural precision, perfect for scenes with dramatic tone
π¨ Recommended Settings
π Suggested LoRA weight: 0.6 - 0.9 (higher values may lead to detail clutter)
π Trigger keywords: midjourneyv7,midjourney style
π Compatible models: FLUX.1 dev
π Recommended sampler: euler
π Recommended generation settings: 25 steps, 1024Γ1024 resolution or higher (hires.fix suggested for upscaling to 1280px or above)
π CFG scale: 3.5 ~ 6.5 (extreme values may destabilize the style)
π Training Info
Trained on a custom dataset of 500 images generated using Midjourney, covering a wide variety of themes including environments, characters, architecture, and objects
Multi-round fine-tuning was performed to improve edge sharpness, color consistency, and structural fidelity
Special emphasis was placed on capturing Midjourneyβs signature lighting and compositional aesthetics
π License & Usage
πΉ For non-commercial use only (CC BY-NC)
πΉ For commercial licensing, please contact the creator via CivitAI direct message


