Updated: Nov 27, 2025
styleVerified: 18 days ago
SafeTensor
Created on Civitai
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.
-Trained On Over 150 Seedream 4.0 Images.
-Dataset Includes But Is Not Limited To: Photoreal People And Scenes, Animals/Mythical Creatures, Dynamic Lighting Shots, 3D Animation, Abstract Surrealism/Geometry, And SOME Anime (Minimal).
-Trained On 1024x1024 Resolution For The Highest Quality Renders Possible Without Upscaling.
TRIGGERWORD: S33DR34M
CFG Scale: 1.0 (GGUF)
Sampling Steps: 20 - 30
Sampler: Euler
Recommended Strength: .5 - 1.0
( be warned, sometimes the lora can be TOO strong when using a strength of 1.0, causing slight morphing or hallucination. try running on the same seed and prompt and decrease the strength to .8 to see if the issue resolves. )
Creator Notes: the LoRA excels in photorealism as well as abstract surrealism. Testing the lora in comfy provided some astonishing results as far as fidelity and quality. The dynamic lighting training really pushes the model to render strong dynamic light sources and contrast well onto characters/scenes. I highly recommend this lora for close-up portraits or abstract ideas youve been trying with other loras. This essentially removed my need for my stack i currently personally use and i expect it will for you. :)
**Works EXTREMELY well when paired with the DyPE Node, Link to the DyPE workflow below:
https://civitai.com/models/2087354/rebels-dype-image-upscaler
(DyPE node is in custom nodes manager, just search "DyPE" and it can be downloaded right in ComfyUI. My workflow is just condensed for LOW VRAM users!)