{ "name": "SmoothMix Animations LTX", "slug": "smoothmix-animations-ltx", "base_model": "LTX-2.3", "media_type": "video", "model_id": 2524245, "version_id": 2837052, "version_name": "v1.0", "source": "civitai", "url": "https://civitai.com/models/2524245", "repo_id": null, "file_path": null, "author": "DigitalPastel", "category": { "key": "style", "label": "Visual style" }, "modalities": [ "t2v", "i2v" ], "primary_modality": "t2v", "checkpoint_family": "dev", "trigger_words": [ "smoothmixanime", "anime style", "smoothmixrealism", "realistic style" ], "recommended_settings": { "LoRA strength": "1.0" }, "chaining": { "author": { "stacks": "not stated", "notes": "", "avoid": [], "requires": [], "order": "", "strength": "", "stated": false }, "general_rules": [ "Stacking an acceleration LoRA with a task LoRA is officially sanctioned: the ID-LoRA template chains LoraLoaderModelOnly(ltx_2.3_22b_distilled_1.1_lora_dynamic_fro09_avg_rank_111_bf16, 0.5) into LoraLoaderModelOnly(ltx-2.3-id-lora-talkvid-3k, 1.0) on the dev checkpoint.", "The Python CLI exposes `--lora [strength]` as a repeatable flag with default strength 1.0, so multiple adapters at once is a supported configuration, not a hack.", "Load the acceleration/distilled LoRA on the dev checkpoint only. The two-stage pipelines require it for the full model but explicitly do not use it for DistilledPipeline, ICLoraPipeline or DubItPipeline, which already run distilled weights.", "Do not stack two distillation LoRAs (e.g. the official rank-384 / rank-111 one plus a community DMD LoRA). The DaSiWa DMD LoRA already blends 75-80% of the official distilled delta into its audio-output route and is described by its author as recreating the distilled model on top of stock dev, i.e. a replacement.", "IC-LoRAs are not chained like ordinary LoRAs: each carries a reference_downscale_factor in its safetensors metadata and needs a node that reads it (GetICLoRAParameters -> LTXVAddGuide in core, or LTXICLoRALoaderModelOnly + LTXAddVideoICLoRAGuide in ComfyUI-LTXVideo). Without one the factor silently defaults to 1.", "Two IC-LoRA reference guides can be driven from a single IC-LoRA: the CrossView Warp author wires both a depth-warp video and the original video into separate reference guides and sets latent_downscale_factor = 1 on both.", "DubItPipeline accepts exactly one Dub-It IC-LoRA (`--lora` exactly once) and applies it in both stages — that path is single-adapter by design.", "Community convention on strengths when stacking: the IC-LoRA Dual-Character author recommends 0.6-1.0 for a content LoRA used alone, dropped to 0.3-0.5 when combined with others.", "Some LoRA pairs are designed to be blended against each other — the LTX 2.3 Crisp Enhance / Soft Enhance pair is described by its author as \"can be used together and balanced at different strengths\".", "The gemma-3-12b-it-abliterated LoRA is not a video LoRA and does not belong in the DiT chain — it patches the Gemma text encoder through a model+clip LoraLoader, as the official templates wire it.", "The full, flexible, trainable bf16 model — ltx-2.3-22b-dev.safetensors is 46.15 GB bf16, 29.15 GB as ltx-2.3-22b-dev-fp8.safetensors. This is what LoRAs are trained against, and what four of the six official ComfyUI LTX-2.3 templates load (t2v, i2v, ia2v, id_lora). Needs either real guidance (Lightricks' own single-stage workflow sets LTXVScheduler to 15 steps with MultimodalGuider) or the distill" ] }, "pros": [ "Well established — 3,733 downloads. [source: CivitAI stats]", "Positively received — 281 thumbs-up, 1 down. [source: CivitAI stats]", "Explicit trigger word(s) — smoothmixanime, anime style, smoothmixrealism, realistic style — so the effect can be turned on and off from the prompt. [source: model metadata]", "Author documents use across several modes (T2V, I2V). [source: description]", "Author published concrete recommended settings. [source: description]" ], "cons": [ "Large at 1286 MB — slower to load and heavier to stack. [source: file size]" ], "file": { "name": "SmoothMix_Animations_LTX2.safetensors", "size_bytes": 1348027424, "size_mb": 1285.58, "sha256": "02a9e8d1884db23e022828f7d7221d5c93fc3c8d7924a022d8e4ab0a81ebb31f" }, "stats": { "downloads": 3733, "thumbs_up": 281, "thumbs_down": 1 }, "popularity": { "downloads": 3733, "thumbs_up": 281, "thumbs_down": 1, "comments": 0, "approval": "100% positive (281/282)", "age_days": 135.9, "downloads_per_day": 27.47, "rank_by_downloads": 8, "rank_by_momentum": 7, "of_total": 150, "adoption_tier": "Top tier", "momentum": "Hot" }, "selection_score": 7505.0, "selection_reasons": [ "3,733 downloads", "281 thumbs-up", "documented triggers + usage notes", "top-ranked for its category" ], "example_media": [ { "file": "01_video.mp4", "type": "video", "resolution": "704x1280", "prompt": "", "poster": "01_poster.jpg" }, { "file": "02_video.mp4", "type": "video", "resolution": "704x1280", "prompt": "", "poster": "02_poster.jpg" }, { "file": "03_video.mp4", "type": "video", "resolution": "704x1280", "prompt": "", "poster": "03_poster.jpg" }, { "file": "04_video.mp4", "type": "video", "resolution": "960x960", "prompt": "", "poster": "04_poster.jpg" }, { "file": "05_video.mp4", "type": "video", "resolution": "768x1152", "prompt": "", "poster": "05_poster.jpg" }, { "file": "06_video.mp4", "type": "video", "resolution": "768x1152", "prompt": "", "poster": "06_poster.jpg" } ], "tags": [ "style" ], "scraped_at": "2026-09-05T00:10:48+00:00", "mirror": { "repo_id": "Alex995647/loras-ltxv-2.3", "path_in_repo": "smoothmix-animations-ltx", "url": "https://huggingface.co/Alex995647/loras-ltxv-2.3/tree/main/smoothmix-animations-ltx", "local_weights_removed": true, "bytes_freed": 1348027424 }, "path": "ltxv-2.3/smoothmix-animations-ltx", "archive": { "comments_total": 16, "comments": [ { "author": "7526757", "score": 11, "text": "EDIT: Effect is quite strong. Tricky to tune down. Because LTX-2.3 is only one model versus Wan's double pass, it's hard to use this supplementally - the look wants to be pretty prominent.\n\nLooking forward to trying this. Your Wan LoRA is one of the best. But holy hell have I had a rough time with LTX-2.3. When it works, it's neat, but so many duff generations. Hit rate is like... 1 in 10? 1 in 20? Anywho, this looks like a nice addition to the quiver.", "created_at": "2026-04-07T05:51:54.604Z" }, { "author": "10580073", "score": 16, "text": "I had to rent a rtx pro 6000 on runpod to be able to train my loras. If I trained locally it was going to take 48 HOURS to do 1 epoch. Runpod took 6 hours. Ltx takes ALOT to train it hit 94gb of VRAM and like 130gb of ram use on runpod I do not have pockets deep enough for a rig like that running locally lmao. From my experience ltx needs alot of videos. My motion lora took 155 videos that were all top tier quality and VBVR took 4800. The paper for VBVR says they kept getting improvements until they hit like 200k videos but that runpod bill would put me in generational debt.", "created_at": "2026-04-07T22:52:56.205Z" }, { "author": "879720", "score": 0, "text": "Thanks for your hard work. I haven't tried it yet.\n\nWould you consider making a ko fi support page at some point?", "created_at": "2026-04-22T04:49:30.124Z" }, { "author": "11926876", "score": 0, "text": "30 hours, lol this looks great i clearly need to rethink because 30 hours is basically just starting ._.", "created_at": "2026-04-22T06:30:38.192Z" }, { "author": "2874261", "score": 0, "text": "The LTX 2.3 is still very raw. The sound seems to have been improved. But otherwise, it's terrible.", "created_at": "2026-04-24T12:39:17.294Z" }, { "author": "10564", "score": 2, "text": "Would be nice if you could retrain this on eros/sulphur. It should be much easier since it already knows nsfw concepts so it only needs to learn the animation/render style.", "created_at": "2026-06-26T22:25:30.094Z" }, { "author": "98528", "score": 0, "text": "thank you for this LTX2.3 version. but i'm very disappointed, it's far of the quality lvl as the Wan22 loras. Especially for the penis.\nAs i make almost only futa video, i can said, the shape of the penis in the ltx 2.3 lora is just horrible. Especially the glans, and testicules.\nIn the smoothmixXXX Wan22 lora, it's just amazing, but not with this one.", "created_at": "2026-07-15T09:44:08.308Z" }, { "author": "73787", "score": 0, "text": "why do you train a style LORA with activation words?", "created_at": "2026-07-17T11:22:24.649Z" } ], "faq": [ { "question": "What is SmoothMix Animations LTX?", "answer": "SmoothMix Animations LTX is a style LoRA designed for LTX Video-family checkpoints, published by [DigitalPastel](/users/DigitalPastel) on CivitAI in April 2026. It layers on top of any compatible LTX Video checkpoint. You load the checkpoint first, then apply this LoRA on top of it. It has attracted 3,734 downloads. The model is flagged NSFW on CivitAI and can produce explicit content." } ], "trigger_words": [ "smoothmixanime", "anime style", "smoothmixrealism", "realistic style" ], "tags": [ "style" ], "rating": 0, "rating_count": 0, "favorite_count": 0, "creator": "DigitalPastel", "url": "https://civarchive.com/models/2524245?modelVersionId=2837052" } }