---
base_model: Wan-AI/Wan2.1-I2V-14B-480P-Diffusers
tags:
- lora
- qlora
- egocentric
- exo-to-ego
- video-generation
- egoX
---

# EgoX mix LoRA — rank-128 QLoRA (single 24 GB GPU reproduction)

A **LoRA adapter** for **Wan2.1-I2V-14B-480P** reproducing **EgoX** (exocentric → egocentric video
generation) on a **single 24 GB GPU** (vs the paper's 4× H200). Trained with the paper's
Geometry-Guided Self-Attention (GGA), scaled down to fit the hardware budget.

![joker in-the-wild](joker_mix_ckpt1984_infer.gif)

*exo input | our vitS ego-prior | ego pred (our prior) | ego pred (shipped prior)*

## Training
- **Adapter:** LoRA rank 128 / α 128 on the Wan-14B transformer (NF4 4-bit base + 8-bit AdamW)
- **Data:** 552 clips, **all 17 EgoX domains** (Ego-Exo4D); 495 train / 57 val
- **Resolution:** 49×176×704 · **LR** 2e-5, constant-with-warmup · effective batch 4
- **16 epochs / 1,984 steps**, held-out val loss **0.247 → 0.173** (checkpoint-1984)

## Use
Load as an unfused PEFT adapter onto an NF4-quantized Wan2.1-I2V-14B with GGA enabled
(see the [reproduction repo](https://github.com/skr3178/egoX): `EgoX/infer_nf4.py --use_GGA`).

## Caveats
Scaled-down repro: 176-res output, NF4-r128, 552 clips (vs the paper's ~3,510). Image-criteria
scores trail the paper (~11 vs 16 PSNR) due to resolution + budget, not pipeline fidelity.