Updated: Jun 3, 2025
styleVerified: a month ago
SafeTensor
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.
*It's not perfect but will be useful*
FILMV4.introduction
FilmV4 Version is Training on 1440p high res image
V4 lora-gen ratios support expansion will be from 4:3 to 2.39:1
get better colors and lighting, and more film realistic style
V4 has undergone special method deal with De-Artifacts-blur version
deal with artifacts and blur, we will provide more efficient workflows to achieve the best results with V4 usage
In order to avoid many unnecessary hassles, if you want to use V4 For commercial use, please check out the model terms
CHECK OUT>About this Version (get the workflow) Weight recommend: 0.5-0.7(FP32)
____________________________________________________________________________________
Optimize training and aesthetic evaluation based on (V1_80+|V2_300+|V3_600+|V4_???) high-quality training sets in 2.35:1/2.39:1 size(Prefer cool tones)
FLUX LORA WEIGHT: 0.7 (0.4 adapt with other lora together)
sampler size: 1024x432/1504x640/1920x816(best size for result with high vram)
sampler step: 30-50steps
sampler recommend: dpm_adaptive+simple/beta(best)<<res3m+beta57
Flux Guidance: 2.5(best)-3.5cfg
____________________________________________________________________________________
All Reviews and feedback are welcome
PS:(model is not limited to fixed proportions and can optimize composition and aesthetics and add the appropriate cinematic feel at any size)
Tag prompts can be triggered(In the Trigger Words of model details)< (FILM_V3)
____________________________________________________________________________________
Tag List:
1.SIZE: Close Up, Extreme Wide size
2.LEN: Wide len, Long len, Medium len, Ultra Wide len, Telephoto len
3.SHOT: Low angle shot, High angle shot, clean single shot, Aerial shot, Establishing shot,
2 shot, Over the shoulder shot, Insert shot
4.Composition: Balanced composition, Center Composition, Right Heavy Composition, Left Heavy composition, Symmetrical composition
5.Light: Soft light, Hard light, Top light, Under light, Side light, Edge light, backlight, Sithouette
6.Contrast: High contrast, Low contrast
PROMPT-TAG example:"Extreme Wide size, Wide len, High angle Clean single shot, balanced composition, Hard light, High contrast"
____________________________________________________________________________________
Model terms: The film style model designed for open source film_art research
will not be used for any commercial disputes arising from this model.
Any results generated From the model in the process will be borne by you personally.