theblindeman/twha | Run with an API on Replicate

theblindeman / twha

Cold

Public
3 runs
H100

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

prompt

*string

Shift + Return to add a new line

twha . wearing an unbuttoned men's dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.twha . wearing an unbuttoned men's dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.

Text prompt for video generation

negative_prompt

string

Shift + Return to add a new line

Things you do not want to see in your video

Default: ""

image

file

Image to use as a starting frame for image to video generation.

aspect_ratio

string

The aspect ratio of the video. 16:9, 9:16, 1:1, etc.

Default: "16:9"

frames

integer

The number of frames to generate (1 to 5 seconds)

Default: 81

resolution

string

The resolution of the video. 720p is not supported for 1.3b.

Default: "480p"

lora_strength_model

number

Strength of the LORA applied to the model. 0.0 is no LORA.

Default: 1

lora_strength_clip

number

Strength of the LORA applied to the CLIP model. 0.0 is no LORA.

Default: 1

fast_mode

string

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.

Default: "Balanced"

sample_steps

integer

(minimum: 1, maximum: 60)

Number of generation steps. Fewer steps means faster generation, at the expensive of output quality. 30 steps is sufficient for most prompts

Default: 30

sample_guide_scale

number

(minimum: 0, maximum: 10)

Higher guide scale makes prompt adherence better, but can reduce variation

Default: 5

sample_shift

number

(minimum: 0, maximum: 10)

Sample shift factor

Default: 8

seed

integer

Set a seed for reproducibility. Random by default.

Run this model in Node.js with one line of code:

npx create-replicate --model=theblindeman/twha

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";
import fs from "node:fs";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run theblindeman/twha using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "theblindeman/twha:f72ba92f2bd5924059ed3d85470461f8ee733cef7092540650dea8ef7407a740",
  {
    input: {
      frames: 81,
      prompt: "twha . wearing an unbuttoned men's dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.",
      fast_mode: "Balanced",
      resolution: "720p",
      aspect_ratio: "16:9",
      sample_shift: 8,
      sample_steps: 30,
      negative_prompt: "",
      lora_strength_clip: 1,
      sample_guide_scale: 5,
      lora_strength_model: 1
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run theblindeman/twha using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "theblindeman/twha:f72ba92f2bd5924059ed3d85470461f8ee733cef7092540650dea8ef7407a740",
    input={
        "frames": 81,
        "prompt": "twha . wearing an unbuttoned men's dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.",
        "fast_mode": "Balanced",
        "resolution": "720p",
        "aspect_ratio": "16:9",
        "sample_shift": 8,
        "sample_steps": 30,
        "negative_prompt": "",
        "lora_strength_clip": 1,
        "sample_guide_scale": 5,
        "lora_strength_model": 1
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run theblindeman/twha using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "theblindeman/twha:f72ba92f2bd5924059ed3d85470461f8ee733cef7092540650dea8ef7407a740",
    "input": {
      "frames": 81,
      "prompt": "twha . wearing an unbuttoned men\'s dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.",
      "fast_mode": "Balanced",
      "resolution": "720p",
      "aspect_ratio": "16:9",
      "sample_shift": 8,
      "sample_steps": 30,
      "negative_prompt": "",
      "lora_strength_clip": 1,
      "sample_guide_scale": 5,
      "lora_strength_model": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

{
  "completed_at": "2025-05-29T18:23:05.421872Z",
  "created_at": "2025-05-29T18:08:10.213000Z",
  "data_removed": false,
  "error": null,
  "id": "yvjb9bfbwnrma0cq3mxarcv7xr",
  "input": {
    "frames": 81,
    "prompt": "twha . wearing an unbuttoned men's dress shirt and panties is walking through a kitchen. she holds her morning coffee mug in her hands. as she walks the unbuttoned shirt sways with her movement, teasing the viewer with a glimpse of her breasts. the scene is almost like a fantasy you have in a dream of a woman you want to be with. set in a modern cozy kitchen.",
    "fast_mode": "Balanced",
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "sample_shift": 8,
    "sample_steps": 30,
    "negative_prompt": "",
    "lora_strength_clip": 1,
    "sample_guide_scale": 5,
    "lora_strength_model": 1
  },
  "logs": "Random seed set to: 4210953954\n2025-05-29T18:12:03Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmpge8xnzyg/weights url=https://replicate.delivery/xezq/UAq7jOCkueVxHauHA4K0m4jCDxj5ylcGem5MANylM1QQjkxUA/trained_model.tar\n2025-05-29T18:12:03Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local\n2025-05-29T18:12:06Z | INFO  | [ Complete ] dest=/tmp/tmpge8xnzyg/weights size=\"307 MB\" total_elapsed=3.326s url=https://replicate.delivery/xezq/UAq7jOCkueVxHauHA4K0m4jCDxj5ylcGem5MANylM1QQjkxUA/trained_model.tar\nChecking inputs\n====================================\nChecking weights\n✅ wan_2.1_vae.safetensors exists in ComfyUI/models/vae\n✅ wan2.1_t2v_14B_bf16.safetensors exists in ComfyUI/models/diffusion_models\n✅ 14b_933cfa912fe6e85b8f11146de130994c.safetensors exists in loras directory\n✅ umt5_xxl_fp16.safetensors exists in ComfyUI/models/text_encoders\n====================================\nRunning workflow\n[ComfyUI] got prompt\nExecuting node 40, title: EmptyHunyuanLatentVideo, class type: EmptyHunyuanLatentVideo\nExecuting node 49, title: Load LoRA, class type: LoraLoader\n[ComfyUI] Requested to load WanTEModel\nExecuting node 6, title: CLIP Text Encode (Positive Prompt), class type: CLIPTextEncode\n[ComfyUI] loaded completely 113273.66620521546 10835.4765625 True\nExecuting node 48, title: ModelSamplingSD3, class type: ModelSamplingSD3\nExecuting node 3, title: KSampler, class type: KSampler\n[ComfyUI] Requested to load WAN21\n[ComfyUI] loaded completely 115028.28298866272 27251.406372070312 True\n[ComfyUI]\n[ComfyUI] 0%|          | 0/30 [00:00<?, ?it/s]\n[ComfyUI] Resetting TeaCache state\n[ComfyUI]\n[ComfyUI] 3%|▎         | 1/30 [00:28<13:57, 28.89s/it]\n[ComfyUI] 7%|▋         | 2/30 [01:05<15:40, 33.58s/it]\n[ComfyUI] 10%|█         | 3/30 [01:42<15:47, 35.09s/it]\n[ComfyUI] TeaCache: Initialized\n[ComfyUI]\n[ComfyUI] 13%|█▎        | 4/30 [02:28<17:04, 39.40s/it]\n[ComfyUI] 17%|█▋        | 5/30 [02:28<10:31, 25.26s/it]\n[ComfyUI] 20%|██        | 6/30 [03:06<11:46, 29.42s/it]\n[ComfyUI] 23%|██▎       | 7/30 [03:06<07:36, 19.85s/it]\n[ComfyUI] 27%|██▋       | 8/30 [03:44<09:20, 25.47s/it]\n[ComfyUI] 30%|███       | 9/30 [03:44<06:08, 17.56s/it]\n[ComfyUI] 33%|███▎      | 10/30 [04:21<07:54, 23.72s/it]\n[ComfyUI] 37%|███▋      | 11/30 [04:21<05:13, 16.51s/it]\n[ComfyUI] 40%|████      | 12/30 [04:59<06:54, 23.01s/it]\n[ComfyUI] 43%|████▎     | 13/30 [04:59<04:33, 16.09s/it]\n[ComfyUI] 47%|████▋     | 14/30 [05:37<06:02, 22.63s/it]\n[ComfyUI] 50%|█████     | 15/30 [05:37<03:57, 15.86s/it]\n[ComfyUI] 53%|█████▎    | 16/30 [06:15<05:13, 22.36s/it]\n[ComfyUI] 57%|█████▋    | 17/30 [06:15<03:23, 15.68s/it]\n[ComfyUI] 60%|██████    | 18/30 [06:52<04:26, 22.24s/it]\n[ComfyUI] 63%|██████▎   | 19/30 [06:53<02:51, 15.60s/it]\n[ComfyUI] 67%|██████▋   | 20/30 [07:31<03:45, 22.56s/it]\n[ComfyUI] 70%|███████   | 21/30 [07:32<02:23, 15.94s/it]\n[ComfyUI] 73%|███████▎  | 22/30 [08:09<02:59, 22.41s/it]\n[ComfyUI] 77%|███████▋  | 23/30 [08:09<01:50, 15.73s/it]\n[ComfyUI] 80%|████████  | 24/30 [08:47<02:13, 22.26s/it]\n[ComfyUI] 83%|████████▎ | 25/30 [08:47<01:18, 15.63s/it]\n[ComfyUI] 87%|████████▋ | 26/30 [09:26<01:30, 22.56s/it]\n[ComfyUI] 90%|█████████ | 27/30 [09:26<00:47, 15.91s/it]\n[ComfyUI] 93%|█████████▎| 28/30 [10:04<00:44, 22.40s/it]\n[ComfyUI] 97%|█████████▋| 29/30 [10:04<00:15, 15.72s/it]\n[ComfyUI] 100%|██████████| 30/30 [10:41<00:00, 22.27s/it]\nExecuting node 8, title: VAE Decode, class type: VAEDecode\nExecuting node 50, title: Video Combine 🎥🅥🅗🅢, class type: VHS_VideoCombine\n[ComfyUI] 100%|██████████| 30/30 [10:41<00:00, 21.40s/it]\n[ComfyUI] Prompt executed in 658.70 seconds\noutputs:  {'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'}]}}\n====================================\nR8_Wan_00001.png\nR8_Wan_00001.mp4",
  "metrics": {
    "predict_time": 662.239612717,
    "total_time": 895.208872
  },
  "output": [
    "https://replicate.delivery/xezq/6WcdL4ttNTp5LVikLFkBgO7j4URkfZ4AgMhBiFkWdK6EwyYKA/R8_Wan_00001.mp4"
  ],
  "started_at": "2025-05-29T18:12:03.182259Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/bcwr-ak7d6hzqsfitwf7g3q3ohtrjbxdo32bldk23bo5s5ffu5rsshfka",
    "get": "https://api.replicate.com/v1/predictions/yvjb9bfbwnrma0cq3mxarcv7xr",
    "cancel": "https://api.replicate.com/v1/predictions/yvjb9bfbwnrma0cq3mxarcv7xr/cancel"
  },
  "version": "f72ba92f2bd5924059ed3d85470461f8ee733cef7092540650dea8ef7407a740"
}

Generated in

11 minutes

Tweak it ShareReport View full prediction

Random seed set to: 4210953954
2025-05-29T18:12:03Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmpge8xnzyg/weights url=https://replicate.delivery/xezq/UAq7jOCkueVxHauHA4K0m4jCDxj5ylcGem5MANylM1QQjkxUA/trained_model.tar
2025-05-29T18:12:03Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local
2025-05-29T18:12:06Z | INFO  | [ Complete ] dest=/tmp/tmpge8xnzyg/weights size="307 MB" total_elapsed=3.326s url=https://replicate.delivery/xezq/UAq7jOCkueVxHauHA4K0m4jCDxj5ylcGem5MANylM1QQjkxUA/trained_model.tar
Checking inputs
====================================
Checking weights
✅ wan_2.1_vae.safetensors exists in ComfyUI/models/vae
✅ wan2.1_t2v_14B_bf16.safetensors exists in ComfyUI/models/diffusion_models
✅ 14b_933cfa912fe6e85b8f11146de130994c.safetensors exists in loras directory
✅ umt5_xxl_fp16.safetensors exists in ComfyUI/models/text_encoders
====================================
Running workflow
[ComfyUI] got prompt
Executing node 40, title: EmptyHunyuanLatentVideo, class type: EmptyHunyuanLatentVideo
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 113273.66620521546 10835.4765625 True
Executing node 48, title: ModelSamplingSD3, class type: ModelSamplingSD3
Executing node 3, title: KSampler, class type: KSampler
[ComfyUI] Requested to load WAN21
[ComfyUI] loaded completely 115028.28298866272 27251.406372070312 True
[ComfyUI]
[ComfyUI] 0%|          | 0/30 [00:00<?, ?it/s]
[ComfyUI] Resetting TeaCache state
[ComfyUI]
[ComfyUI] 3%|▎         | 1/30 [00:28<13:57, 28.89s/it]
[ComfyUI] 7%|▋         | 2/30 [01:05<15:40, 33.58s/it]
[ComfyUI] 10%|█         | 3/30 [01:42<15:47, 35.09s/it]
[ComfyUI] TeaCache: Initialized
[ComfyUI]
[ComfyUI] 13%|█▎        | 4/30 [02:28<17:04, 39.40s/it]
[ComfyUI] 17%|█▋        | 5/30 [02:28<10:31, 25.26s/it]
[ComfyUI] 20%|██        | 6/30 [03:06<11:46, 29.42s/it]
[ComfyUI] 23%|██▎       | 7/30 [03:06<07:36, 19.85s/it]
[ComfyUI] 27%|██▋       | 8/30 [03:44<09:20, 25.47s/it]
[ComfyUI] 30%|███       | 9/30 [03:44<06:08, 17.56s/it]
[ComfyUI] 33%|███▎      | 10/30 [04:21<07:54, 23.72s/it]
[ComfyUI] 37%|███▋      | 11/30 [04:21<05:13, 16.51s/it]
[ComfyUI] 40%|████      | 12/30 [04:59<06:54, 23.01s/it]
[ComfyUI] 43%|████▎     | 13/30 [04:59<04:33, 16.09s/it]
[ComfyUI] 47%|████▋     | 14/30 [05:37<06:02, 22.63s/it]
[ComfyUI] 50%|█████     | 15/30 [05:37<03:57, 15.86s/it]
[ComfyUI] 53%|█████▎    | 16/30 [06:15<05:13, 22.36s/it]
[ComfyUI] 57%|█████▋    | 17/30 [06:15<03:23, 15.68s/it]
[ComfyUI] 60%|██████    | 18/30 [06:52<04:26, 22.24s/it]
[ComfyUI] 63%|██████▎   | 19/30 [06:53<02:51, 15.60s/it]
[ComfyUI] 67%|██████▋   | 20/30 [07:31<03:45, 22.56s/it]
[ComfyUI] 70%|███████   | 21/30 [07:32<02:23, 15.94s/it]
[ComfyUI] 73%|███████▎  | 22/30 [08:09<02:59, 22.41s/it]
[ComfyUI] 77%|███████▋  | 23/30 [08:09<01:50, 15.73s/it]
[ComfyUI] 80%|████████  | 24/30 [08:47<02:13, 22.26s/it]
[ComfyUI] 83%|████████▎ | 25/30 [08:47<01:18, 15.63s/it]
[ComfyUI] 87%|████████▋ | 26/30 [09:26<01:30, 22.56s/it]
[ComfyUI] 90%|█████████ | 27/30 [09:26<00:47, 15.91s/it]
[ComfyUI] 93%|█████████▎| 28/30 [10:04<00:44, 22.40s/it]
[ComfyUI] 97%|█████████▋| 29/30 [10:04<00:15, 15.72s/it]
[ComfyUI] 100%|██████████| 30/30 [10:41<00:00, 22.27s/it]
Executing node 8, title: VAE Decode, class type: VAEDecode
Executing node 50, title: Video Combine 🎥🅥🅗🅢, class type: VHS_VideoCombine
[ComfyUI] 100%|██████████| 30/30 [10:41<00:00, 21.40s/it]
[ComfyUI] Prompt executed in 658.70 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

Examples

View more examples

Run time and cost

This model runs on Nvidia H100 GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

This model doesn't have a readme.