andreasjansson/wan-1.3b-inpaint 🔢📝🖼️✓ → 🖼️
About
Inpainting and video2video experiments with Wan 2.1

Example Output
Prompt:
"A robot walking on the street"
Output
Performance Metrics
410.26s
Prediction Time
410.27s
Total Time
All Input Parameters
{ "seed": -1, "prompt": "A robot walking on the street", "strength": 0.9, "mask_video": "https://replicate.delivery/xezq/N94eQyo3WBVWXyN8Aj4kpwQ42l6prb578m9tCJWaFgREnlJKA/masked_video.mp4", "expand_mask": 10, "guide_scale": 5, "input_video": "https://replicate.delivery/xezq/UpqkYwl7hTqTJ5rfIE0eb5WpDv6quG6PeFcdtH157Y2uonloA/tmp2dc62s6m.output.mp4", "sampling_steps": 30, "negative_prompt": "", "frames_per_second": 16, "inpaint_fixup_steps": 0 }
Input Parameters
- seed
- Random seed. Leave blank for random
- prompt (required)
- Prompt for inpainting the masked area
- strength
- Strength of inpainting effect, 1.0 is full regeneration
- mask_video
- Mask video (white areas will be inpainted). Leave blank for video-to-video
- expand_mask
- Expand the mask by a number of pixels
- guide_scale
- Guidance scale for prompt adherence
- input_video (required)
- Original video to be inpainted
- sampling_steps
- Number of sampling steps
- negative_prompt
- Negative prompt
- frames_per_second
- Output video FPS
- keep_aspect_ratio
- Keep the aspect ratio of the input video. This will degrade the quality of the inpainting.
- inpaint_fixup_steps
- Number of steps for final inpaint fixup. Ignored when in video-to-video mode (when mask_video is empty)
Output Schema
Output
Example Execution Logs
Using seed: 3692929301 Loaded 141 frames from /tmp/tmpywa_qk39tmp2dc62s6m.output.mp4 Loaded 141 frames from /tmp/tmpthdg8haymasked_video.mp4 0%| | 0/27 [00:00<?, ?it/s] 4%|▎ | 1/27 [00:14<06:15, 14.44s/it] 7%|▋ | 2/27 [00:28<06:01, 14.45s/it] 11%|█ | 3/27 [00:43<05:47, 14.46s/it] 15%|█▍ | 4/27 [00:57<05:32, 14.46s/it] 19%|█▊ | 5/27 [01:12<05:18, 14.46s/it] 22%|██▏ | 6/27 [01:26<05:03, 14.47s/it] 26%|██▌ | 7/27 [01:41<04:49, 14.47s/it] 30%|██▉ | 8/27 [01:55<04:34, 14.46s/it] 33%|███▎ | 9/27 [02:10<04:20, 14.47s/it] 37%|███▋ | 10/27 [02:24<04:05, 14.47s/it] 41%|████ | 11/27 [02:39<03:51, 14.47s/it] 44%|████▍ | 12/27 [02:53<03:37, 14.47s/it] 48%|████▊ | 13/27 [03:08<03:22, 14.47s/it] 52%|█████▏ | 14/27 [03:22<03:08, 14.47s/it] 56%|█████▌ | 15/27 [03:36<02:53, 14.47s/it] 59%|█████▉ | 16/27 [03:51<02:39, 14.47s/it] 63%|██████▎ | 17/27 [04:05<02:24, 14.47s/it] 67%|██████▋ | 18/27 [04:20<02:10, 14.47s/it] 70%|███████ | 19/27 [04:34<01:55, 14.47s/it] 74%|███████▍ | 20/27 [04:49<01:41, 14.47s/it] 78%|███████▊ | 21/27 [05:03<01:26, 14.47s/it] 81%|████████▏ | 22/27 [05:18<01:12, 14.47s/it] 85%|████████▌ | 23/27 [05:32<00:57, 14.46s/it] 89%|████████▉ | 24/27 [05:47<00:43, 14.46s/it] 93%|█████████▎| 25/27 [06:01<00:28, 14.46s/it] 96%|█████████▋| 26/27 [06:16<00:14, 14.46s/it] 100%|██████████| 27/27 [06:30<00:00, 14.46s/it] 100%|██████████| 27/27 [06:30<00:00, 14.46s/it] Moviepy - Building video output.mp4. Moviepy - Writing video output.mp4 t: 0%| | 0/141 [00:00<?, ?it/s, now=None] t: 8%|▊ | 11/141 [00:00<00:01, 105.39it/s, now=None] t: 16%|█▌ | 22/141 [00:00<00:01, 104.08it/s, now=None] t: 23%|██▎ | 33/141 [00:00<00:01, 103.61it/s, now=None] t: 31%|███ | 44/141 [00:00<00:00, 103.81it/s, now=None] t: 39%|███▉ | 55/141 [00:00<00:00, 100.08it/s, now=None] t: 47%|████▋ | 66/141 [00:00<00:00, 93.21it/s, now=None] t: 55%|█████▍ | 77/141 [00:00<00:00, 95.89it/s, now=None] t: 62%|██████▏ | 87/141 [00:00<00:00, 93.66it/s, now=None] t: 69%|██████▉ | 97/141 [00:01<00:00, 93.10it/s, now=None] t: 76%|███████▌ | 107/141 [00:01<00:00, 90.07it/s, now=None] t: 83%|████████▎ | 117/141 [00:01<00:00, 86.44it/s, now=None] t: 89%|████████▉ | 126/141 [00:01<00:00, 87.00it/s, now=None] t: 96%|█████████▌| 135/141 [00:01<00:00, 84.94it/s, now=None] Moviepy - Done ! Moviepy - video ready output.mp4 Saved video to output.mp4
Version Details
- Version ID
7abfdb3370aba087f9a5eb8b733c2174bc873a957e5c2c4835767247287dbf89
- Version Created
- March 10, 2025