zsxkib/allegro | Run with an API on Replicate

Powerful text-to-video model that generates high-quality videos up to 6 seconds at 15 FPS and 720p resolution from simple text prompt

Cold

Public

144 runs

A100 (80GB)

License

GitHub

Weights

Paper

Playground API Examples README Versions

Input

user_prompt

*string

Shift + Return to add a new line

Text prompt for video generation

guidance_scale

number

(minimum: 0, maximum: 20)

Guidance scale

Default: 7.5

num_sampling_steps

integer

(minimum: 10, maximum: 100)

Number of sampling steps

Default: 20

enable_cpu_offload

boolean

Enable CPU offload (reduces GPU memory usage but increases inference time)

Default: false

num_frames

integer

(minimum: 16, maximum: 128)

Number of frames to generate

Default: 88

height

integer

(minimum: 256, maximum: 1024)

Height of the generated video

Default: 720

width

integer

(minimum: 256, maximum: 1920)

Width of the generated video

Default: 1280

fps

integer

(minimum: 1, maximum: 60)

Frames per second for the output video

Default: 15

seed

integer

Random seed (set to None for random)

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

npx create-replicate --model=zsxkib/allegro

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

const output = await replicate.run(
  "zsxkib/allegro:a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919",
  {
    input: {
      fps: 15,
      width: 1280,
      height: 720,
      num_frames: 88,
      user_prompt: "a baby riding on the back of a dog",
      guidance_scale: 7.5,
      enable_cpu_offload: false,
      num_sampling_steps: 20
    }
  }
);

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

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

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

output = replicate.run(
    "zsxkib/allegro:a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919",
    input={
        "fps": 15,
        "width": 1280,
        "height": 720,
        "num_frames": 88,
        "user_prompt": "a baby riding on the back of a dog",
        "guidance_scale": 7.5,
        "enable_cpu_offload": False,
        "num_sampling_steps": 20
    }
)

# To access the file URL:
print(output.url())
#=> "http://example.com"

# To write the file to disk:
with open("my-image.png", "wb") as file:
    file.write(output.read())

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 zsxkib/allegro 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": "zsxkib/allegro:a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919",
    "input": {
      "fps": 15,
      "width": 1280,
      "height": 720,
      "num_frames": 88,
      "user_prompt": "a baby riding on the back of a dog",
      "guidance_scale": 7.5,
      "enable_cpu_offload": false,
      "num_sampling_steps": 20
    }
  }' \
  https://api.replicate.com/v1/predictions

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

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/zsxkib/allegro@sha256:a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919 \
  -i 'fps=15' \
  -i 'width=1280' \
  -i 'height=720' \
  -i 'num_frames=88' \
  -i 'user_prompt="a baby riding on the back of a dog"' \
  -i 'guidance_scale=7.5' \
  -i 'enable_cpu_offload=false' \
  -i 'num_sampling_steps=20'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/zsxkib/allegro@sha256:a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "fps": 15,
      "width": 1280,
      "height": 720,
      "num_frames": 88,
      "user_prompt": "a baby riding on the back of a dog",
      "guidance_scale": 7.5,
      "enable_cpu_offload": false,
      "num_sampling_steps": 20
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

{
  "completed_at": "2024-10-23T16:14:15.185147Z",
  "created_at": "2024-10-23T16:03:16.391000Z",
  "data_removed": false,
  "error": null,
  "id": "xswwaz1v4xrj00cjq8f910r75r",
  "input": {
    "fps": 15,
    "width": 1280,
    "height": 720,
    "num_frames": 88,
    "user_prompt": "a baby riding on the back of a dog",
    "guidance_scale": 7.5,
    "enable_cpu_offload": false,
    "num_sampling_steps": 20
  },
  "logs": "Using seed: 28725\n  0%|          | 0/20 [00:00<?, ?it/s]\n  5%|▌         | 1/20 [00:24<07:37, 24.06s/it]\n 10%|█         | 2/20 [00:39<05:42, 19.01s/it]\n 15%|█▌        | 3/20 [01:04<06:07, 21.65s/it]\n 20%|██        | 4/20 [01:29<06:06, 22.90s/it]\n 25%|██▌       | 5/20 [01:53<05:54, 23.60s/it]\n 30%|███       | 6/20 [02:18<05:36, 24.02s/it]\n 35%|███▌      | 7/20 [02:43<05:15, 24.29s/it]\n 40%|████      | 8/20 [03:08<04:53, 24.47s/it]\n 45%|████▌     | 9/20 [03:33<04:30, 24.59s/it]\n 50%|█████     | 10/20 [03:58<04:06, 24.67s/it]\n 55%|█████▌    | 11/20 [04:23<03:42, 24.72s/it]\n 60%|██████    | 12/20 [04:47<03:18, 24.76s/it]\n 65%|██████▌   | 13/20 [05:12<02:53, 24.78s/it]\n 70%|███████   | 14/20 [05:37<02:28, 24.80s/it]\n 75%|███████▌  | 15/20 [06:02<02:04, 24.81s/it]\n 80%|████████  | 16/20 [06:27<01:39, 24.82s/it]\n 85%|████████▌ | 17/20 [06:52<01:14, 24.83s/it]\n 90%|█████████ | 18/20 [07:16<00:49, 24.83s/it]\n 95%|█████████▌| 19/20 [07:41<00:24, 24.83s/it]\n100%|██████████| 20/20 [08:06<00:00, 24.81s/it]\n100%|██████████| 20/20 [08:06<00:00, 24.33s/it]",
  "metrics": {
    "predict_time": 578.186704536,
    "total_time": 658.794147
  },
  "output": "https://replicate.delivery/yhqm/ZyxTCVLA0lKgEV0YeSVlee3Ms72iXRBxQ2oveAfjrbv7apNdC/output.mp4",
  "started_at": "2024-10-23T16:04:36.998443Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://stream.replicate.com/v1/files/qoxq-y6m3otr6qzwhr5j7fzsna5hmuy44yxyxuwptlse5grjt6dyqhg6a",
    "get": "https://api.replicate.com/v1/predictions/xswwaz1v4xrj00cjq8f910r75r",
    "cancel": "https://api.replicate.com/v1/predictions/xswwaz1v4xrj00cjq8f910r75r/cancel"
  },
  "version": "a3922c1ba8e826531492ad08dc2759ce225d5bba48153ca3b5849ef1a4399919"
}

Generated in

9 minutes 38 seconds

Tweak it Iterate in playground ShareReport View full prediction

Examples

View more examples

Run time and cost

This model runs on Nvidia A100 (80GB) GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

Allegro is a powerful text-to-video model that generates high-quality videos up to 6 seconds at 15 FPS and 720p resolution from simple text input.

kwaivgi/kling-v1.6-standard

Generate short videos from text prompts. Optionally condition on a start image to create image-to-video clips and include up to 4 reference images as scene elements. Choose 5s or 10s duration with 720p output at 30fps, and set aspect ratio to 16:9, 9:16, or 1:1. Supports negative prompts to steer content and returns a video.

1.4m runs

Official

kwaivgi/kling-v2.1

Generate short videos from a start image and a text prompt. Produce 5 or 10 second clips at 24 fps in 720p (standard) or 1080p (pro). Optionally supply an end image in pro mode to guide the final frame or interpolate between start and end.

2.5m runs

Official

bytedance/seedance-1-lite

Generate short videos from text prompts or a starting image. Produce 2–12 second clips at 24 fps in up to 1080p resolution across aspect ratios including 16:9, 4:3, 1:1, 3:4, 9:16, 21:9, and 9:21. Guide subjects, style, and multi-character interactions with 1–4 reference images for character, clothing, and environment consistency. Optionally lock the camera, set a random seed for reproducibility, and anchor start/end frames with first- and last-frame images. Outputs a video.

1.6m runs

Official

kwaivgi/kling-v2.5-turbo-pro

Generate 5–10 second videos from text prompts or a single starting image. Accept a required prompt and optional first-frame image, and output short clips with fluid motion, stable frames, and coherent pacing. Preserve color, lighting, and mood across frames with refined conditioning, and follow multi-step, causal instructions for complex camera moves. Suited for marketing assets, creator shorts, film/animation previz, and educational explainers.

740.3k runs

Official

google/veo-3-fast

Generate videos with sound from a text prompt or from a reference image plus prompt. Create 4–8 second clips at 720p or 1080p in 16:9 or 9:16, with optional native audio generation. Run fast and cost-efficiently for text-to-video storytelling, product spins, concept shots, and image-to-video animations.

128.3k runs

Official

kwaivgi/kling-v1.6-pro

Generate 5–10 second 1080p videos from a text prompt. Provide a start or end image to anchor the first or last frame (one is required) and optionally add up to four reference images as scene elements. Select 16:9, 9:16, or 1:1 aspect ratios and refine results with negative prompts.

759.1k runs

Official

kwaivgi/kling-v2.0

Generate 5–10 second 720p videos from a text prompt. Optionally animate from a starting image by using it as the first frame (image-to-video). Select aspect ratios 16:9, 9:16, or 1:1. Outputs silent video.

79.2k runs

Official

wavespeedai/wan-2.1-t2v-480p

Generate 480p videos from a text prompt with accelerated inference. Accept a prompt plus optional LoRA weights and scaling to apply styles or characters, with controls for aspect ratio (16:9, 9:16), seed for reproducibility, negative prompt, guidance scale, sampling steps, and flow shift. Include a safety checker with an option to disable it. Output is a silent video.

179.4k runs

Official

Input

Output

Examples

Run time and cost

Readme

Related models

kwaivgi/kling-v1.6-standard

kwaivgi/kling-v2.1

bytedance/seedance-1-lite

kwaivgi/kling-v2.5-turbo-pro

google/veo-3-fast

kwaivgi/kling-v1.6-pro

kwaivgi/kling-v2.0

wavespeedai/wan-2.1-t2v-480p