{ "name": "CheekPinch Pals LORA V1.0", "slug": "cheekpinch-pals-lora-v1-0", "base_model": "Wan 2.2", "media_type": "video", "model_id": 1908220, "version_id": 2159837, "version_name": "CheekPinch Pals LORA", "source": "civitai", "url": "https://civitai.com/models/1908220", "repo_id": null, "file_path": null, "author": "HUPOHUPO", "category": { "key": "detail", "label": "Detail / anatomy fixer" }, "modalities": [ "i2v" ], "primary_modality": "i2v", "checkpoint_family": "i2v-a14b", "trigger_words": [], "recommended_settings": { "Sampler": "euler" }, "chaining": { "author": { "stacks": "not stated", "notes": "", "avoid": [], "requires": [], "order": "", "strength": "", "stated": false }, "general_rules": [ "The high/low pair is not a stack. Every A14B LoRA that ships two files needs the high-noise file on the high-noise UNET chain and the low-noise file on the low-noise UNET chain — two LoraLoaderModelOnly nodes on two different models, never both on one model.", "Never load a high-noise LoRA onto the low-noise expert or vice versa. It loads without a single warning because the two checkpoints are architecturally identical, and the damage shows up only as degraded output. Filename is the only guard.", "A single-file A14B LoRA with no high/low marking (common for Wan2.1-era LoRAs and for LoRAs extracted from merged models) goes on BOTH experts at the same strength. Loading it only on the low-noise expert is the safer half-measure when it fights the composition.", "Stack subject/style LoRAs after the accelerator LoRA in the chain: UNETLoader -> Lightning LoRA -> subject LoRA -> ModelSamplingSD3. Order within LoraLoaderModelOnly chains is additive so it rarely matters mathematically, but keeping the accelerator first matches every reference workflow.", "Do not stack two step-distillation LoRAs on the same expert. One accelerator per expert.", "When adding subject LoRAs on top of a 4-step distill LoRA, reduce the subject LoRA strength rather than the distill LoRA strength — the distill LoRA is what keeps 4 steps viable at cfg 1.0.", "Keep the LoRA strength for the two experts independently tunable. Because the high-noise expert owns layout and the low-noise expert owns detail, the community routinely runs a style LoRA weaker on high-noise and full-strength on low-noise.", "TI2V-5B LoRAs chain like any single-model LoRA and can never be mixed into an A14B chain, or the reverse — the two have entirely different tensor geometry." ] }, "pros": [ "Well established — 1,310 downloads. [source: CivitAI stats]", "Positively received — 149 thumbs-up with no down-votes. [source: CivitAI stats]", "Compact at 293 MB — cheap to stack with other LoRAs. [source: file size]", "Author published concrete recommended settings. [source: description]" ], "cons": [ "No trigger word — the LoRA is always on once loaded, so strength is the only control. [source: model metadata]" ], "file": { "name": "CheekPinch Pals LORA.safetensors", "size_bytes": 306807976, "size_mb": 292.59, "sha256": "1b2a399a7e73efc5167aa9f87a5be4f06c404aae3213a48a7ac03fe48183a8ac" }, "stats": { "downloads": 1310, "thumbs_up": 149, "thumbs_down": 0 }, "popularity": { "downloads": 1310, "thumbs_up": 149, "thumbs_down": 0, "comments": 3, "approval": "100% positive (149/149)", "age_days": 371.9, "downloads_per_day": 3.52, "rank_by_downloads": 129, "rank_by_momentum": 95, "of_total": 150, "adoption_tier": "Niche", "momentum": "Long tail" }, "selection_score": 54.8, "selection_reasons": [ "1,310 downloads", "149 thumbs-up", "top-ranked for its category" ], "example_media": [ { "file": "01_video.mp4", "type": "video", "resolution": "1440x2560", "prompt": "", "poster": "01_poster.jpg" }, { "file": "02_video.mp4", "type": "video", "resolution": "1440x2560", "prompt": "", "poster": "02_poster.jpg" }, { "file": "03_video.mp4", "type": "video", "resolution": "1440x2560", "prompt": "", "poster": "03_poster.jpg" }, { "file": "04_video.mp4", "type": "video", "resolution": "1440x2560", "prompt": "", "poster": "04_poster.jpg" }, { "file": "05_video.mp4", "type": "video", "resolution": "1440x2560", "prompt": "", "poster": "05_poster.jpg" }, { "file": "06_video.mp4", "type": "video", "resolution": "480x832", "prompt": "", "poster": "06_poster.jpg" } ], "tags": [ "poses", "wan video 2.2 i2v-a14b", "lora" ], "scraped_at": "2026-09-05T03:23:02+00:00", "mirror": { "repo_id": "Alex995647/loras-wan-2.2", "path_in_repo": "cheekpinch-pals-lora-v1-0", "url": "https://huggingface.co/Alex995647/loras-wan-2.2/tree/main/cheekpinch-pals-lora-v1-0", "local_weights_removed": true, "bytes_freed": 306807976 }, "path": "wan-2.2/cheekpinch-pals-lora-v1-0", "archive": { "comments_total": 4, "comments": [ { "author": "1850477", "score": 5, "text": "👶Baby Cheeks👶with Adults? How dare you?!!😆", "created_at": "2025-08-29T06:42:32.530Z" }, { "author": "1101998", "score": 0, "text": "I guess it works better if only i HIgh Noise. I tested with a normal WAN 2.2 workflow and the end video became a bit damaged.", "created_at": "2025-09-02T21:17:33.494Z" }, { "author": "9496666", "score": 0, "text": "Hello, could I ask you something? I noticed that one of your LoRAs no longer appears, and it was quite interesting. I wanted to ask if that LoRa was removed for any reason, or if there’s a way to download it. The specific LoRa I’m referring to is \"Splash-and-Transform\".", "created_at": "2025-09-16T20:32:30.974Z" } ], "faq": [ { "question": "What is CheekPinch Pals LORA V1.0?", "answer": "CheekPinch Pals LORA V1.0 is a pose LoRA designed for Wan Video-family checkpoints, published by [HUPOHUPO](/users/HUPOHUPO) on CivitAI in August 2025. It layers on top of any compatible Wan Video checkpoint. You load the checkpoint first, then apply this LoRA on top of it. Wan 2.2 uses a two-pass architecture — a high-noise model handles motion and scene composition, and a low-noise model refines detail. Both models need to be loaded together for a complete pipeline; a single pass on its own won't produce usable output. When using speed LoRAs (lightx2v, CausVid, Lightning), the community consensus is to apply them only on the low-noise pass and keep the high-noise pass unmodified — otherwise you lose most of the motion and lighting quality that makes Wan 2.2 worth using. It has attracted 1,311 downloads. It's intended for wan video 2.2 i2v-a14b and lora content." } ], "trigger_words": [], "tags": [ "wan video 2.2 i2v-a14b", "poses", "lora" ], "rating": 0, "rating_count": 0, "favorite_count": 0, "creator": "HUPOHUPO", "url": "https://civarchive.com/models/1908220?modelVersionId=2159837" } }