This is my first Lora  :)

An audio and video LoRA for LTX-2.3 / Eros. It generates its own sound.

Trigger: BEANFLK — all caps, first word of your prompt.

QUICK START

Base model: Eros v1.4 (LTX-2.3 22B)
Text encoder: Gemma 3 12B IT
Loader: LTX2 LoRA Loader Advanced
Strength: 1.4–1.69 general, 1.75 for reliable insertion
Branch weights: video 1, audio 1, v2a 1, a2v 1, other 1
Resolution: 800x800, match your source image's aspect
Duration: 10s @ 25fps
Speed LoRA: ltx-2.3-22b-distilled-lora-384-1.1 at 0.5 — keep it on

STRENGTH — READ THIS BIT

1.0 — too weak. Her hand hovers just off the skin. No real contact, and she doesn't read as enjoying it.

1.4 to 1.69 — the general working band. Clean rubbing with real contact and expression.

1.75 — reliable insertion, but only with a properly primed source image. See below.

Above ~1.69 — artifacts creep in:
- Nipples wiggle and jiggle faster than the breast itself. Not catastrophic, but disconcerting once you see it.
- On a flat-chested subject, her breasts start to grow.

It's a gradient, not a wall. Plenty of renders above 1.69 are fine — just know what you're looking for.

YOUR SOURCE IMAGE MATTERS AS MUCH AS STRENGTH

For insertion, her hand must be at least directly above the opening. Any further away and insertion is not guaranteed — and cranking the strength will not rescue it. You'll get surface rubbing at a higher strength plus the artifacts that come with it.

Fix the frame, not the slider.

Rubbing has no such requirement. It works from almost any starting pose.

WHAT IT DOES

A general-purpose masturbation model, not a single-trick LoRA.

Audio — the standout. Base Eros gives gritty, choppy sound. This gives clean continuous wet sounds, breathing, moans and a real climax.
Expression — noticeably more emotion than the base model, which just moves a hand around.
Rubbing — clean and convincing in the working band.
Finger insertion — needs 1.75 and a primed source image.
Fluids — supported, but be specific about what the fluid is doing and where it's coming from. Vague prompting doesn't work.

PROMPTING

You are not limited to what it was trained on. This LoRA sits on top of a capable base model. Ask for things outside its training — licking her fingers, changing position, whatever — and the base handles it while the LoRA supplies the action and the sound.

Put BEANFLK first, then describe the scene and action in plain prose. Keep the audio description in the same sentence flow rather than a separate block.

Example:
BEANFLK pale woman, tight crop on her pussy and hand, pushing her fingertips deep into her vagina, wet sounds of her pussy with moans and sighs

Keep the word "deep" if you want depth. It's deliberately absent from the training captions, so the base model supplies it and the two don't fight.

SAMPLER SETTINGS

Pass 1: euler_cfg_pp, sigmas 0.85, 0.4219, 0.0
Pass 2: euler_ancestral_cfg_pp, sigmas 1.0, 0.9, 0.72, 0.4, 0.0
CFG 1.2 both passes
Upscaler: ltx-2.3-spatial-upscaler-x2-1.1

Don't set the speed LoRA to 0. Those 2-step and 4-step schedules only work because it's engaged. Turn it off without raising your step count and you get blurry, vibrating motion.