CivArchive
    LTX2.5_Stubelius Remix [NSFW] - Beta1
    NSFW

    LTX 2.5 Stubelius Remix — beta 2 out now!


    Int8 merged with Distilled lora so you can run it on its own + other finetunes.


    LTX 2.5 Stubelius Remix — beta 1

    A custom LTX 2.5 checkpoint that brings the 2.3-era stylized/NSFW lineage onto the 2.5 engine — built because existing 2.5 ports lean hard into photoreal humans and lose the 2D/stylized soul.

    Method (short version): task-vector delta arithmetic, not a naive merge. Deltas extracted from Sulphur 2 (concept & composition) and 10Eros (2D/stylized prior) against their LTX 2.3 base, then injected into the LTX 2.5 transformer with per-block-depth and per-component weighting — early blocks preserve composition knowledge, late blocks favor the stylized prior, photoreal MLP contributions deliberately suppressed. 4,349 tensors, 48 blocks.

    Usage

    • Load as a diffusion model (UNETLoader / Load Diffusion Model) with your normal LTX 2.5 VAEs + text encoder

    • Run with the LTX distilled LoRA at 0.6–1.0 strength, 10–12 steps. Without it, output is blurry/underbaked (standard 2.5-dev behavior)

    • CFG 1 with the distill LoRA; your usual samplers (euler / linear_quadratic tested)

    • Fully compatible with the Stubelius LTX Director pack (seed hunt, refine, Licon MSR references, IC-LoRA, ingredients)

    What it's for

    2D / 2.5D / anime-styled characters on the full LTX 2.5 feature set (motion, audio, dialogue). NSFW-capable across styles via the Sulphur/Eros lineage. Beta 1: style balance is a first pass and will evolve — feedback steers beta 2.

    Credits

    • TenStrip — Sulphur 2 and 10Eros, the models this stands on

    • Lightricks — LTX 2.5 base (LTX-Video community license applies)

    • Merge method & balance — Stubelius

    Adult-capable model — intended for consenting-adult creative work only. Check the base and donor model licenses before commercial use.

    Description

    FAQ

    Comments (12)

    MysteriousString420508Sep 5, 2026· 7 reactions
    CivitAI

    16gb peeasant here, I'd love to try an int8/fp8 version !

    not downloading 40gb to then convert to then try out. nobody feels like it, seeing from the content posted.

    Stuubzzz
    Author
    Sep 7, 2026· 1 reaction

    I will be uploading the beta2 (once finished) as int8.

    herkus_baronas631Sep 5, 2026· 6 reactions
    CivitAI

    How about int8/fp8?

    @spacejumper2021824 Im not programmer. Im just user. And i dont have 5090 or a6000.

    @herkus_baronas631 I recommend it because it's a very easy operation; even if you only have a GTX 4060, you can complete the conversion in under a minute, eliminating the need to wait for someone else to upload. If you're interested, you can try using Codex or DeepSeek Harness to help you complete the process.

    @spacejumper2021824 Oh, thanks.

    Stuubzzz
    Author
    Sep 7, 2026· 4 reactions

    I'm doing the int8 for beta2. This first beta I'm keeping at bf16

    Dumcluck51Sep 17, 2026

    I have a 4080 16GB GPU on Ubuntu Linux. I followed the link to the comfy-model-tools but I have no clue how to implement it. Does it need a special venv with dedicated python environment? I have python integrated with Comfyui in that venv - could I somehow use that. In short - there's a huge gap between people who are savvy with coding tools and environments and those of us who are "just users".

    TheWizzardSep 19, 2026
    CivitAI

    fp8?

    Checkpoint
    LTXV 2.5

    Details

    Downloads
    372
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/5/2026
    Updated
    9/23/2026
    Deleted
    -

    Files