MUZED Motion LoRA — LTX-2 19B
Motion LoRA for contemporary dance video generation on
LTX-2 19B . Trained on real dancer footage to capture fluid upper-body and full-body movement patterns.
Quick Start
1 # Using LTX-2 inference pipeline
2 python -m ltx_pipelines.ti2vid_one_stage \
3 --checkpoint-path /path/to/ltx-2-19b-dev.safetensors \
4 --gemma-root /path/to/gemma-text-encoder \
5 --lora /path/to/lora_weights_step_01500.safetensors 1.0 \
6 --prompt "DNCMOV upper body dance movement, fluid arm extension with gentle torso rotation, contemporary dance" \
7 --negative-prompt "worst quality, inconsistent motion, blurry, jittery, distorted" \
8 --width 1536 --height 864 --num-frames 161 --frame-rate 30 \
9 --num-inference-steps 50 --video-cfg-guidance-scale 4.0 \
10 --output-path output.mp4
With first-frame conditioning (I2V)
1 # Add --image flag for image-to-video
2 --image first_frame.png 0 1.0
Trigger Token
DNCMOV
Training Details
Parameter Value Base model LTX-2 19B (ltx-2-19b-dev.safetensors) Training mode LoRA Rank 64 Alpha 64 Target modules attn1.* + attn2.* + ff.net.* (video-only)Resolution 1536×864×161 (27,216 tokens) Frame rate 30 fps Dataset 125 clips × 5.37s from 3 source videos Steps 1500 (ongoing to 2000) Learning rate 1e-4 (linear decay) Batch size 1 (grad accumulation 2) Mixed precision bf16 Gradient checkpointing enabled First-frame conditioning 50% (supports both T2V and I2V) Hardware 1×H200 (141GB) Training time ~12.4h for 1500 steps (~29.8s/step) Final loss 0.229
Target Modules (video-only attention + FFN)
attn1.to_k, attn1.to_q, attn1.to_v, attn1.to_out.0
attn2.to_k, attn2.to_q, attn2.to_v, attn2.to_out.0
ff.net.0.proj, ff.net.2Dataset
125 clips at 1536×864, 161 frames (5.37s) each, 30fps. Source: 3 real dancer videos — 2 upper-body (canon), 1 full-body. Captions describe movement type, body parts, direction, and energy using dance vocabulary. Captioned with Gemini 2.0 Flash + Qwen2.5-Omni, manually curated.
Movement vocabulary : flowing, sharp, rhythmic, sustained, percussive, lateral, circular, contracting, expanding, meditative, explosive.
Checkpoints
File Step Loss Notes lora_weights_step_01400.safetensors1400 ~0.23 lora_weights_step_01500.safetensors1500 0.229 End of first run
More checkpoints (1600-2000) will be added as training continues.
Pipeline
Raw video → Slice (161f) → Caption (Gemini) → Curate → VAE encode → Train LoRA
↓
Generate (T2V / I2V) ← Load LoRA + base modelRecommended Inference Settings
Parameter Value Resolution 1536×864×161 (match training) Guidance (CFG) 3.0–5.0 STG scale 1.0 Inference steps 40–50 LoRA strength 0.8–1.0 Frame rate 30 fps Quantization fp8-cast (for GPUs < 80GB)
License
LTX2 Community License
