Sign In

Ultimate Cinema XL

30

240

536

13

Verified:

SafeTensor

Type

LoRA

Stats

64

34

539

Reviews

Published

Dec 21, 2025

Base Model

Flux.1 D

Usage Tips

Strength: 1

Trigger Words

CineM3n

Hash

AutoV2
BC51246B46
default creator card background decoration
Gold Tier Supporter Badge December 2025
Loochi1's Avatar

Loochi1

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.

UPDATE V2

SIGNIFICANT IMPROVEMENTS FROM V1

V2 For sdxl is alot better now, IMPORTANT, FOR V2 ADD: "in the style of CinneM3n" at the end of your prompts, use more natural language prompts insted of tags, use keywords like Cinematic lighting, Dramatic warm orange vibrant lighting / color pallete, backlit silhouette, DOSENT NEED NEGATIVE PROMPT, at least from what iv tested, if you get bad faces try describing the face and subject more

p.s Male focsed training images, i havent tried generateing females, In next update ill add more backrounds and add females

V2-SDXL EXAMPLES:


A cinematic, film-like nighttime portrait shows a young person leaning over a bridge, looking back toward the camera with a quiet, introspective expression as the city glows behind them. Dramatic, bright, vibrant orange streetlights contrast sharply with a teal-blue night sky and deep shadows, while soft bokeh, shallow depth of field, and subtle analog grain give the image a moody, nostalgic, almost 35mm film aesthetic in the style of CinneM3n

Negative:

NONE

For comfyUI use:
Dpmpp_2m_sde karras, 30-40 steps, 4-7 cfg 1024px and 1152px

V1 SDXL + FLUX

IMPORTANT: Male focused lora, i havent tested females but it should work, use “CineM3n” at the start of your prompts, add stuff like cinematic lighting, color grading etc for a better effect. For sdxl you may need to use face fix and a hand fix negative embedding, it was trained on relatively short prompts.

This is only Version 1 and trained with about 50 images it will most likely have some artefacts

Example:

Positive

CineM3n, a cinematic photo of a young man standing still behind rising flames at dusk, his figure partially silhouetted against a stormy sky, with dramatic cinematic backlighting, high contrast, heavy film grain, and moody warm–cool color grading that makes the fire glow violently against the darkened horizon like a climactic final scene, old-school style

NEGATIVE (recomended)

unaestheticXL_cbp62, worst quality, low quality, lowres, blurry, out of focus, noisy, grainy, washed out, overexposed, underexposed, bad anatomy, bad proportions, malformed body, deformed, disfigured, extra fingers, missing fingers, fused fingers, extra limbs, missing limbs, poorly drawn hands, poorly drawn face, NEGATIVE_HANDS

use unaestheticXL https://civitai.com/models/119032?modelVersionId=797474

and for hands https://civitai.com/models/583583?modelVersionId=651084

P.s works great with base sdxl haven’t tried it with others yet

If you use on site gen use DPM2 25-40 steps 5-6 cfg