by MiniMax

Silent 6 or 10-second clips from a still at 768p, or 6 seconds at 1080p

minimax/hailuo-2.3/image-to-video
  • Up to 1080p
  • 6–10s clips
  • Image to Video

About MiniMax Hailuo 2.3

Silent 6 or 10-second clips from a still at 768p, or 6 seconds at 1080p.

Call it with the model name minimax/hailuo-2.3/image-to-video on POST /jobs/createTask, from $0.163.

Introduction

MiniMax released Hailuo 2.3 on October 28, 2025 as the next step after Hailuo 02, together with a lighter Hailuo 2.3 Fast. This endpoint exposes both as image-to-video: you upload the frame the shot should start on, describe what happens next, and receive a silent MP4. The generation_mode field chooses between the full model and the Fast variant without changing anything else about the request.

Key features

Motion that reads as physical

MiniMax built 2.3 around fluid body movement, steadier response to camera moves and more convincing light and shadow. In our 1080p test a dancer finishing a spin kept the red silk dress swinging after she stopped, and the hem settled instead of freezing in place.

Faces that change gradually

Micro-expressions are the other headline of this release. A clip that starts on a neutral look can move through a glance, a breath and a smile within six seconds; our rainy-window test shows a slow smile building while the camera pushes in.

Two models, one request shape

standard and fast take the same image, prompt, resolution and length, so you can draft a shot in Fast and rerender the chosen prompt in standard without rewriting anything. Output follows the opening image's proportions: a 16:9 still came back as 1364 by 768 at 24 frames per second.

Request parameters

ParameterRequiredDescription
duration_secondsNoLength of the generated clip. 6 · 10 · Default: 6
generation_modeNoChoose a mode from the available options. It may affect cost, speed and quality. standard · fast · Default: standard
image_urlYesThe picture to start from.
promptYesDescribe what you want to make.
resolutionNoChoose an available output size. 768p · 1080p · Default: 768p

Pricing

Prices are in USD and charged to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.

OptionUSD per run
generation_mode fast$0.163
generation_mode standard$0.256
duration_seconds 10 · generation_mode fast$0.264
generation_mode fast · resolution 1080p$0.266
generation_mode standard · resolution 1080p$0.448
duration_seconds 10 · generation_mode standard$0.512

Send the same request body to POST /jobs/quote to see the estimated and maximum USD charge before running. A quote does not run the task or charge your balance.

Use cases

Moving key art

Campaign stills, cover art and poster photography gain a short moving version for feeds and teasers. Light the still the way the first second of the clip should look, because the model starts exactly there.

Dance, fashion and fabric

Spins, walks and garments moving around a body are the strongest material for this model. Start from a frame already in motion and say how the move ends, so the clip has somewhere to land.

Draft reels at volume

When you need many variations of one shot, render them in Fast at 768p, compare, and spend standard mode or 1080p only on the version that goes out.

Prompt guide & pro tips

One action, then the camera

Write the prompt as a short shot description: who moves, what they do, how it ends, and what the camera does. For example, have the woman turn from the window, tuck her hair behind her ear and smile while the camera pushes in. Several unrelated actions in six seconds tend to blur together.

Respect the frame you uploaded

The opening image fixes costume, lighting and composition. Describe changes that can grow from it, such as a turn, a step or a gust of wind, rather than asking for a new outfit or location the first frame does not show.

Plan length and size together

10-second clips are available at 768p only; 1080p is 6 seconds.

API quick start

Call the MiniMax Hailuo 2.3 API in three steps

  1. Create an API key. Sign in to PopNow and create a key in the API console. Keys start with sk-popnow-. API calls are billed in USD from your PopNow balance.
  2. Send the request. POST /jobs/createTask with model minimax/hailuo-2.3/image-to-video and an input object that follows the parameter table. The response returns a task ID straight away.
  3. Get the result. Poll GET /jobs/recordInfo with the task ID, or pass a callBackUrl and we post the result to it. Download the output links when the task succeeds.
# 1. Create the task. Keep the Idempotency-Key if you retry after a network error.
export IDEMPOTENCY_KEY="$(uuidgen)"
curl -X POST https://api.popnow.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  -d '{
  "model": "minimax/hailuo-2.3/image-to-video",
  "input": {
    "prompt": "A slow aerial shot drifting over a misty pine forest at sunrise",
    "image_url": "https://media.popnow.ai/influencer/examples/marco.webp",
    "generation_mode": "standard",
    "resolution": "768p",
    "duration_seconds": 6
  }
}'

# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $POPNOW_API_KEY"

Replace the prompt and any links with your own. The API tab has the response, polling and callback reference.

MiniMax Hailuo 2.3 API questions

How much does the MiniMax Hailuo 2.3 API cost?

MiniMax Hailuo 2.3 costs $0.163 per run. Prices by option: fast $0.163 per run, standard $0.256 per run, 10 · fast $0.264 per run, fast · 1080p $0.266 per run. API usage is charged in USD to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.

What is the model name for MiniMax Hailuo 2.3 in the API?

Put minimax/hailuo-2.3/image-to-video for Image to Video in the model field of POST /jobs/createTask.

Can Hailuo 2.3 make a video from text alone?

Not on this endpoint. Every request needs an opening image in image_url, and the prompt describes what happens after that frame. If you only have an idea, create the first frame with an image model, then animate it here.

Does the output keep my image's aspect ratio?

Yes. The clip follows the proportions of the opening image: a 16:9 still came back as 1364 × 768 and a slightly narrower photo as 1344 × 768 at 768p. Crop the still before uploading if you need a specific delivery shape.

Does Hailuo 2.3 generate sound?

No. The files we received carry a video track only. Add music, effects or dialogue in your edit after downloading the clip.

How long does a clip take?

In our runs a Fast 6-second 768p clip took about three minutes, a standard 6-second 768p clip about four, and 1080p or 10-second clips between three and a half and five minutes. Use the task callback rather than holding a connection open.

What are the limits of the MiniMax Hailuo 2.3 API?

Clip length: 6–10 s; Resolutions: 768p · 1080p; Prompt length (characters): 5,000; Largest upload: 10 MB. Requests outside these limits are rejected before anything is charged.

Can I try MiniMax Hailuo 2.3 without writing code?

Run MiniMax Hailuo 2.3 in the Playground with your API USD balance, without an API key. Each run appears in your developer history with its matching code.

All MiniMax models

More about MiniMax Hailuo 2.3