Yume-1.5-5B-720P-MLX

Yume 1.5 5B 720P — MLX

MLX-converted weights for Yume 1.5, a 5B-parameter interactive world generation model. Runs natively on Apple Silicon via mlx-video.

Yume 1.5 is fine-tuned from Wan2.2 TI2V-5B and supports both text-to-video and image-to-video generation.

Quick Start

pip install mlx-video

# Image-to-Video
python -m mlx_video.models.wan_2.generate \
    --model-dir ckurasek/Yume-1.5-5B-720P-MLX \
    --image my_photo.jpg \
    --prompt "A dog runs through the scene" \
    --width 832 --height 480 --num-frames 81 \
    --steps 30 --seed 42 \
    --output-path output.mp4

# Text-to-Video (omit --image)
python -m mlx_video.models.wan_2.generate \
    --model-dir ckurasek/Yume-1.5-5B-720P-MLX \
    --prompt "A cinematic shot of a mountain landscape at sunset" \
    --width 832 --height 480 --num-frames 81 \
    --steps 30 --seed 42 \
    --output-path output.mp4

The model directory is auto-downloaded from Hugging Face on first use.

Performance (Apple Silicon)

Benchmarked on M5 Pro (48GB unified memory):

Model Details

• Architecture: Wan2.2 TI2V-5B (DiT backbone, T5 text encoder, 3D VAE)
• Parameters: 5B (transformer) + 5.7B (T5) + 72M (VAE)
• Dim: 3072, 30 layers, 24 heads
• VAE: z_dim=48, stride (4, 16, 16)
• Resolution: 720P native (1280x704), works at lower resolutions
• FPS: 24
• Dtype: bfloat16 (transformer, T5), float32 (VAE)

Conversion

These weights were converted from the original PyTorch checkpoint using the mlx-video conversion pipeline:

python -m mlx_video.models.wan_2.convert \
    --checkpoint-dir /path/to/Yume-5B-720P \
    --output-dir ./yume-mlx \
    --dtype bfloat16 --model-version 2.2

Files

Options

Attribution

• Yume 1.5: Paper | GitHub | Project Page
• Wan2.2: GitHub
• mlx-video: GitHub

Citation

@article{mao2025yume15,
  title={Yume-1.5: A Text-Controlled Interactive World Generation Model},
  author={Mao, Xiaofeng and Li, Zhen and Li, Chuanhao and Xu, Xiaojie and Ying, Kaining and He, Tong and Pang, Jiangmiao and Qiao, Yu and Zhang, Kaipeng},
  journal={arXiv preprint arXiv:2512.22096},
  year={2025}
}
@article{mao2025yume,
  title={Yume: An Interactive World Generation Model},
  author={Mao, Xiaofeng and Lin, Shaoheng and Li, Zhen and Li, Chuanhao and Peng, Wenshuo and He, Tong and Pang, Jiangmiao and Chi, Mingmin and Qiao, Yu and Zhang, Kaipeng},
  journal={arXiv preprint arXiv:2507.17744},
  year={2025}
}

License

Apache 2.0 (same as the original Yume model).