aramintak/frosting-land_wan2 🔢🖼️❓📝 → 🖼️

▶️ 129 runs 📅 Apr 2025 ⚙️ Cog 0.13.7

illustration image-to-video lora text-to-video video-lora-training

About

Example Output

Prompt:

"a kaiju monster walking through a city illustration style"

Output

Performance Metrics

876.87s Prediction Time

925.71s Total Time

All Input Parameters

{
  "image": "https://replicate.delivery/pbxt/Mqg8hCbYcvoUxOZ8gV7vUjMaeh9KX5dzp3BHLTFxzlxvtJGt/filename_00162_.png",
  "frames": 81,
  "prompt": "a kaiju monster walking through a city illustration style",
  "fast_mode": "Balanced",
  "resolution": "720p",
  "aspect_ratio": "16:9",
  "sample_shift": 7,
  "sample_steps": 30,
  "negative_prompt": "ugly, messy, grain, film grain, degradation, messed up",
  "lora_strength_clip": 1,
  "sample_guide_scale": 4,
  "lora_strength_model": 0.9
}

Input Parameters

seed Type: integer: Set a seed for reproducibility. Random by default.
image Type: string: Image to use as a starting frame for image to video generation.
frames Default: 81: The number of frames to generate (1 to 5 seconds)
prompt (required) Type: string: Text prompt for video generation
fast_mode Default: Balanced: Speed up generation with different levels of acceleration. Faster modes may degrade quality somewhat. The speedup is dependent on the content, so different videos may see different speedups.
resolution Default: 480p: The resolution of the video. 720p is not supported for 1.3b.
aspect_ratio Default: 16:9: The aspect ratio of the video. 16:9, 9:16, 1:1, etc.
sample_shift Type: numberDefault: 8Range: 0 - 10: Sample shift factor
sample_steps Type: integerDefault: 30Range: 1 - 60: Number of generation steps. Fewer steps means faster generation, at the expensive of output quality. 30 steps is sufficient for most prompts
negative_prompt Type: stringDefault:: Things you do not want to see in your video
replicate_weights Type: string: Replicate LoRA weights to use. Leave blank to use the default weights.
lora_strength_clip Type: numberDefault: 1: Strength of the LORA applied to the CLIP model. 0.0 is no LORA.
sample_guide_scale Type: numberDefault: 5Range: 0 - 10: Higher guide scale makes prompt adherence better, but can reduce variation
lora_strength_model Type: numberDefault: 1: Strength of the LORA applied to the model. 0.0 is no LORA.

Output Schema

Output

Type: array • Items Type: string • Items Format: uri

Example Execution Logs

Random seed set to: 853381153
2025-04-16T20:12:58Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmp69kkmk5l/weights url=https://replicate.delivery/xezq/UOfH00s5GetS2U9Z2elO0d8IhRo2neOc57g8WyuYGDIDbx5RB/trained_model.tar
2025-04-16T20:12:58Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local
2025-04-16T20:13:01Z | INFO  | [ Complete ] dest=/tmp/tmp69kkmk5l/weights size="307 MB" total_elapsed=3.159s url=https://replicate.delivery/xezq/UOfH00s5GetS2U9Z2elO0d8IhRo2neOc57g8WyuYGDIDbx5RB/trained_model.tar
Checking inputs
✅ /tmp/inputs/image.png
====================================
Checking weights
✅ wan_2.1_vae.safetensors exists in ComfyUI/models/vae
✅ 14b_49b48d33a84a18885dc3ebeddac8b186.safetensors exists in loras directory
✅ umt5_xxl_fp16.safetensors exists in ComfyUI/models/text_encoders
✅ clip_vision_h.safetensors exists in ComfyUI/models/clip_vision
⏳ Downloading wan2.1_i2v_720p_14B_bf16.safetensors to ComfyUI/models/diffusion_models
✅ wan2.1_i2v_720p_14B_bf16.safetensors downloaded to ComfyUI/models/diffusion_models in 29.46s, size: 31270.88MB
====================================
Running workflow
[ComfyUI] got prompt
Executing node 39, title: Load VAE, class type: VAELoader
[ComfyUI] Using pytorch attention in VAE
[ComfyUI] Using pytorch attention in VAE
[ComfyUI] VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
Executing node 55, title: Load Image, class type: LoadImage
Executing node 56, title: Width and height for scaling image to ideal resolution 🪴, class type: Width and height for scaling image to ideal resolution 🪴
Executing node 57, title: 🔧 Image Resize, class type: ImageResize+
Executing node 60, title: Load CLIP Vision, class type: CLIPVisionLoader
Executing node 59, title: CLIP Vision Encode, class type: CLIPVisionEncode
[ComfyUI] Requested to load CLIPVisionModelProjection
[ComfyUI] loaded completely 141327.4875 1208.09814453125 True
Executing node 38, title: Load CLIP, class type: CLIPLoader
[ComfyUI] CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
[ComfyUI] Requested to load WanTEModel
Executing node 7, title: CLIP Text Encode (Negative Prompt), class type: CLIPTextEncode
[ComfyUI] loaded completely 139855.3869140625 10835.4765625 True
Executing node 37, title: Load Diffusion Model, class type: UNETLoader
[ComfyUI] model weight dtype torch.float16, manual cast: None
[ComfyUI] model_type FLOW
Executing node 54, title: WanVideo Tea Cache (native), class type: WanVideoTeaCacheKJ
Executing node 49, title: Load LoRA, class type: LoraLoader
[ComfyUI] Requested to load WanTEModel
Executing node 6, title: CLIP Text Encode (Positive Prompt), class type: CLIPTextEncode
[ComfyUI] loaded completely 139853.3869140625 10835.4765625 True
Executing node 58, title: WanImageToVideo, class type: WanImageToVideo
[ComfyUI] Requested to load WanVAE
[ComfyUI] loaded completely 118209.1806602478 242.02829551696777 True
Executing node 48, title: ModelSamplingSD3, class type: ModelSamplingSD3
Executing node 53, title: WanVideo Enhance A Video (native), class type: WanVideoEnhanceAVideoKJ
Executing node 3, title: KSampler, class type: KSampler
[ComfyUI] Requested to load WAN21
[ComfyUI] loaded completely 112525.20673397521 31269.802368164062 True
[ComfyUI]
[ComfyUI] 0%|          | 0/30 [00:00<?, ?it/s]
[ComfyUI] 3%|▎         | 1/30 [00:37<18:17, 37.85s/it]
[ComfyUI] 7%|▋         | 2/30 [01:21<19:15, 41.28s/it]
[ComfyUI] 10%|█         | 3/30 [02:06<19:23, 43.09s/it]
[ComfyUI] TeaCache: Initialized
[ComfyUI]
[ComfyUI] 13%|█▎        | 4/30 [02:58<20:12, 46.62s/it]
[ComfyUI] 17%|█▋        | 5/30 [03:44<19:13, 46.13s/it]
[ComfyUI] 20%|██        | 6/30 [03:44<12:12, 30.50s/it]
[ComfyUI] 23%|██▎       | 7/30 [04:29<13:32, 35.31s/it]
[ComfyUI] 27%|██▋       | 8/30 [04:29<08:50, 24.13s/it]
[ComfyUI] 30%|███       | 9/30 [05:15<10:47, 30.81s/it]
[ComfyUI] 33%|███▎      | 10/30 [05:15<07:07, 21.35s/it]
[ComfyUI] 37%|███▋      | 11/30 [06:00<09:04, 28.68s/it]
[ComfyUI] 40%|████      | 12/30 [06:00<06:00, 20.00s/it]
[ComfyUI] 43%|████▎     | 13/30 [06:45<07:49, 27.64s/it]
[ComfyUI] 47%|████▋     | 14/30 [06:46<05:09, 19.34s/it]
[ComfyUI] 50%|█████     | 15/30 [07:31<06:48, 27.23s/it]
[ComfyUI] 53%|█████▎    | 16/30 [07:31<04:27, 19.09s/it]
[ComfyUI] 57%|█████▋    | 17/30 [08:17<05:50, 26.96s/it]
[ComfyUI] 60%|██████    | 18/30 [08:17<03:46, 18.91s/it]
[ComfyUI] 63%|██████▎   | 19/30 [09:02<04:55, 26.84s/it]
[ComfyUI] 67%|██████▋   | 20/30 [09:02<03:08, 18.83s/it]
[ComfyUI] 70%|███████   | 21/30 [09:02<01:59, 13.24s/it]
[ComfyUI] 73%|███████▎  | 22/30 [09:48<03:03, 22.91s/it]
[ComfyUI] 77%|███████▋  | 23/30 [09:48<01:52, 16.09s/it]
[ComfyUI] 80%|████████  | 24/30 [10:33<02:29, 24.83s/it]
[ComfyUI] 83%|████████▎ | 25/30 [10:33<01:27, 17.43s/it]
[ComfyUI] 87%|████████▋ | 26/30 [11:19<01:43, 25.76s/it]
[ComfyUI] 90%|█████████ | 27/30 [11:19<00:54, 18.08s/it]
[ComfyUI] 93%|█████████▎| 28/30 [12:04<00:52, 26.22s/it]
[ComfyUI] 97%|█████████▋| 29/30 [12:49<00:31, 31.93s/it]
[ComfyUI] 100%|██████████| 30/30 [13:35<00:00, 35.93s/it]
Executing node 8, title: VAE Decode, class type: VAEDecode
Executing node 50, title: Video Combine 🎥🅥🅗🅢, class type: VHS_VideoCombine
[ComfyUI] 100%|██████████| 30/30 [13:35<00:00, 27.17s/it]
[ComfyUI] Prompt executed in 843.77 seconds
outputs:  {'50': {'gifs': [{'filename': 'R8_Wan_00001.mp4', 'subfolder': '', 'type': 'output', 'format': 'video/h264-mp4', 'frame_rate': 16.0, 'workflow': 'R8_Wan_00001.png', 'fullpath': '/tmp/outputs/R8_Wan_00001.mp4'}]}}
====================================
R8_Wan_00001.png
R8_Wan_00001.mp4

Version Details

Version ID: c2d730536e34c591843983323351e09d1012c14d7596df1f640b1da8dbd102e6
Version Created: April 1, 2025

Run on Replicate →