This is my LoRA extraction of the Cosmos-Predict 2.5-2B. Using the <a target="_blank" rel="ugc" href="https://huggingface.co/nvidia/Cosmos-Predict2.5-2B/tree/main/base/pre-trained">Pre-trained </a>and the <a target="_blank" rel="ugc" href="https://huggingface.co/nvidia/Cosmos-Predict2.5-2B/tree/main/base/distilled">Distilled</a> models. Converted from pt to safetensor then cleaned during extraction. Ready to merge.CFG: 1.0 Steps: 8 - 15 Around 12 is good Strength: ~1.0 Sampler: er_sde / euler_ancsetral Scheduler: sgm_uniform / kl_optimal<h3 id="v1-refresh">v1-Refresh</h3>Budgeted 64 dim. Better balance, lower file size. <hr /><h3 id="v1">v1</h3>I have included 3 variations. 32 / 48 / 64. Up to you which you prefer. I would recommend the 32. Its what i used for all the image previews.<hr />I do this for fun and to learn. Likes, comments and reactions are nice to see if my efforts are worth while and posting. I dont really care about the buzz or whatever but drop me a comment and like if you enjoy what Ive done. &lt;3

Cosmos Turbo LoRA (Distilled Extract)

Budgeted 64 dim. Better balance, lower file size.

Files with SHA256: e01ea13d36bc81ba31a9763b2542a07132a2ffcad3163a3ae7f847e156183cfc