{"product_id":"dds-motion-ue5-8-kimodo-plugin","title":"DDS Motion - UE 5.8 Kimodo Plugin","description":"\u003cp\u003e\u003cstrong\u003eDocumentation: https:\/\/darkdojostudios.com\/pages\/dds-motion-kimodo-installation-guide\u003cbr\u003e\u003cbr\u003eDDS Motion\u003c\/strong\u003e is an Unreal Engine 5.8 plugin that lets you author animation by describing it.\u003cbr\u003eIt runs NVIDIA's Kimodo motion model locally, on your own GPU, with no cloud service, no\u003cbr\u003esubscription and no per-generation cost. Your prompts and your animation never leave your\u003cbr\u003emachine.\u003c\/p\u003e\n\u003cp\u003eIt is not a one-shot generator that hands you a clip and wishes you luck. Generation is the\u003cbr\u003efirst step of a real authoring workflow that ends with a retarget-ready animation on your\u003cbr\u003eown character.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat it does\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eWrite it.\u003c\/strong\u003e Describe the motion you want. Build a timeline of several prompts with\u003cbr\u003eindividual durations and the plugin blends between them, so one clip can walk, plant its\u003cbr\u003efeet and stop.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenerate it.\u003c\/strong\u003e Runs locally on your GPU. Control seed, sample count, diffusion steps and\u003cbr\u003ehow strongly the result follows your text against your posed constraints. Generate several\u003cbr\u003esamples at once and pick the best.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003ePose it.\u003c\/strong\u003e AutoPose gives you hand control over the result. Pose bones directly with full\u003cbr\u003eFK and IK, key what you want held, and let generation respect those keys. Make Loop turns a\u003cbr\u003eclip into a seamless cycle.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eGround it\u003c\/strong\u003e. Attach a prop to the hand or place a chair, crate or ledge in the scene, then\u003cbr\u003epose against something real instead of guessing where the character's hands should be.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eShip it.\u003c\/strong\u003e Trim to the frames you want, choose root motion or in place, and export. Then\u003cbr\u003eretarget onto Manny, Quinn, a MetaHuman or your own skeleton.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eProve it.\u003c\/strong\u003e Every generation writes an immutable result asset recording the recipe hash,\u003cbr\u003ethe resolved seed, the model revision and the artifact checksums. You can always show where\u003cbr\u003ean animation came from and reproduce it exactly.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eBuilt for locomotion sets\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003ePrompt timelines, fixed seeds and consistent settings make it practical to build a set of\u003cbr\u003eclips that blend with each other, which is what Motion Matching and blendspace-based\u003cbr\u003elocomotion actually need. Consistency across a set matters more than any single clip, and\u003cbr\u003ethe reproducibility model is designed for exactly that.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eRuns on a normal graphics card\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eKimodo needs roughly 17 GB of video memory to run everything on the GPU. One setting moves\u003cbr\u003eprompt encoding to system RAM and brings measured peak video memory down to about 2 GB.\u003c\/p\u003e\n\u003cp\u003eMeasured on an AMD Radeon RX 7900 XTX under ROCm, through the plugin's own code path:\u003c\/p\u003e\n\u003cp\u003e| Text encoder | Clip | Peak video memory | Time |\u003cbr\u003e|---|---|---|---|\u003cbr\u003e| GPU | 2 seconds, 1 sample | 15.93 GiB | 45 seconds |\u003cbr\u003e| CPU | 2 seconds, 1 sample | 1.82 GiB | 39 to 52 seconds |\u003cbr\u003e| CPU | 6 seconds, 1 sample | 1.84 GiB | 41 seconds |\u003cbr\u003e| CPU | 6 seconds, 4 samples | 2.01 GiB | 51 seconds |\u003c\/p\u003e\n\u003cp\u003eVideo memory barely moves with longer clips or more samples. Tripling the duration and\u003cbr\u003equadrupling the sample count added under 0.2 GiB. The setting costs you time, not quality:\u003cbr\u003ethe motion model is unchanged and still runs on your GPU.\u003c\/p\u003e\n\u003cp\u003eSystem RAM is a separate budget and it does not change between the two paths. Measured peak\u003cbr\u003eRAM was 15.7 GiB with the encoder on CPU and 16.4 GiB with it on the GPU, so plan for 32 GB\u003cbr\u003eregardless of which card you have.\u003c\/p\u003e\n\u003cp\u003eBecause Unreal also needs video memory, plan for about 6 GB as a practical floor and 8 GB\u003cbr\u003efor comfort. These figures were measured on AMD under ROCm. NVIDIA is expected to behave\u003cbr\u003ethe same way but has not yet been measured on NVIDIA hardware.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eExample content included\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eThe plugin ships with finished work you can open immediately: five recipes, five generated\u003cbr\u003eanimations and two exported clips, playing on the included preview body. You can read every\u003cbr\u003eprompt, timeline and pose key before you install anything else. No model, GPU or account\u003cbr\u003erequired just to look.\u003c\/p\u003e\n\u003cp\u003e---\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat you need\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003e- Unreal Engine 5.8, Windows\u003cbr\u003e- A GPU with about 6 GB of video memory or more\u003cbr\u003e- 32 GB of system RAM. This is not a low video memory requirement, it applies whatever your\u003cbr\u003e  card is: measured peak RAM was 15.7 GiB with the text encoder on CPU and 16.4 GiB with it\u003cbr\u003e  on the GPU, because the model weights pass through system memory either way. 16 GB is not\u003cbr\u003e  enough once Windows and Unreal are also resident.\u003cbr\u003e- About 25 GB of disk for model files and the Python environment\u003cbr\u003e- A free Hugging Face account, and Meta's approval for Llama 3, which is used to encode your\u003cbr\u003e  prompt text\u003c\/p\u003e\n\u003cp\u003eThe motion model, text encoder and GPU runtime are \u003cstrong\u003enot\u003c\/strong\u003e redistributed with the plugin.\u003cbr\u003eYou download them yourself under your own accounts and license acceptances. The included\u003cbr\u003einstallation guide walks through every step, including Meta's approval process.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatus\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eDDS Motion is in beta. It is in active development and shipping real work. The included user\u003cbr\u003eand installation guides are complete and current.\u003cbr\u003e\u003c\/p\u003e","brand":"Dark Dojo Studios","offers":[{"title":"Default Title","offer_id":50497732313325,"sku":null,"price":50.0,"currency_code":"USD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0754\/4740\/2733\/files\/DDSMotion1.png?v=1786948188","url":"https:\/\/darkdojostudios.com\/products\/dds-motion-ue5-8-kimodo-plugin","provider":"Dark Dojo Studios","version":"1.0","type":"link"}