MiniMax H3 Max API

Introduction

MiniMax H3 Max Image to video animates an opening frame supplied in image_url into a clip of 5 to 15 whole seconds at 480p or 768p, with sound generated in the same pass. An optional last_image_url fixes the closing frame as well. The opening frame sets the output shape, and a request that leaves the duration out runs for 8 seconds.

Choose a MiniMax H3 Max mode by starting material

Key features

The frame sets subject, wardrobe and shape

There is no aspect-ratio field on this endpoint: the still you hand in decides the frame of the clip. In the laundromat-spin example nothing about the room, the wardrobe or the flat fluorescent light is written down, because the opening frame settled all of it. The prompt only describes the machine going into its spin cycle and how she reacts.

Pin both ends of the shot

last_image_url adds a closing frame, and it can only be sent together with image_url. The frame-to-frame example starts on one composition and lands on another over eight seconds, so the clip cuts cleanly into an edit. The closing frame constrains the ending, not the middle, and two frames that disagree on light or wardrobe produce a visible negotiation between them.

Five-second minimum, two resolutions

Durations run from 5 to 15 whole seconds, with 8 as the default. The shortest request is already a complete beat: the laundromat-spin clip fits the spin, a laugh, a swing of the legs and a glance at the drum into five seconds.

Request parameters

Pricing

Prices are in credits, the same credits you use on the website (1 credit = $0.01). The hold is taken when the task is accepted and settled to the actual charge when it finishes; failed tasks are refunded.

Each task is rounded up to whole credits, so one run at the lowest tier costs 38 credits for a 5-second clip ($0.38).

For an exact number before you run, send the same body to POST /jobs/quote.

Use cases

Tilt from an object to a person

A single unbroken camera move makes the order of reveal the whole brief. The grimoire-to-bare-shoulder example starts on an open book and tilts up to the reader in a high-backed chair, with sigil light from the pages moving across her shoulder, over eight seconds. The image fixes the opening composition, so the prompt spends its words on what the camera finds and when.

Land a clip on a known composition

When an edit needs a clip to end on a specific frame, supply both frames. Prepare the closing still as a continuation of the opening one, then describe the action that connects them. In frame-to-frame that action is a walk toward camera down a marble corridor that stops looking straight down the lens, with hard rim light, a wind machine and a low synth drone.

Bring a locked-off scene to life

A static wide shot can carry several small movements at once. The laundromat-spin prompt asks for a locked-off camera, a drum spinning behind glass and a fluorescent tube that flickers once, alongside the character's reaction, with a quiet mechanical hum in the audio.

Prompt guide & pro tips

Describe change, not the frame

Write what happens after the opening frame and leave appearance to the image. "The washing machine under her shudders into its spin cycle and she bounces slightly with it" names only movement; nothing in it repeats what the still already shows. If a clause describes something visible in the frame, cut it.

Crop and match frames before upload

Crop the opening still to the delivery shape, since the output follows it. When you add a closing frame, keep subject, wardrobe and light continuous between the two stills, and choose a duration long enough for the connecting action. Across a 15-second clip the middle can wander even when both ends are fixed.

API quick start

Call the MiniMax H3 Max 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- and use the same credits as the website.
2. Send the request. POST /jobs/createTask with model minimax/h3-max/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/h3-max/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",
    "duration_seconds": 8,
    "resolution": "768p"
  }
}'

# 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 H3 Max API questions

All MiniMax models

MiniMax H3 adds reference-to-video and a 2K tier. Max is the quicker option when 768p is enough.

• MiniMax H3 APIExpressive clips with audio, up to 2KFrom 28 credits for a 5-second clip
• MiniMax Speech 2.8 HD APIHigh-fidelity speech in 400+ voices with emotion controlFrom 18 credits per 1,000 characters
• MiniMax Music 3.0 APIFinished stereo songs, with vocals or instrumentalFrom 27 credits per run
• MiniMax H3 Singularity LoRA APICommunity-tuned H3 video that loads your own LoRAsFrom 45 credits for a 5-second clip
• MiniMax H3 Max APIFast clips with ambience, effects and musicFrom 38 credits for a 5-second clip
• MiniMax H3 LoRA APIMiniMax H3 video with sound and up to three of your LoRAsFrom 38 credits for a 5-second clip
• MiniMax H3 Image LoRA APIH3 stills and edits with up to three of your LoRAsFrom 7 credits per image
• MiniMax Speech 2.8 Turbo APIFast speech in 400+ voices with emotion controlFrom 11 credits per 1,000 characters

Other video APIs

• Seedance 2.5 APIByteDance's newest: 30-second clips with native audioFrom 56 credits for a 5-second clip
• Kling 3.0 API4K clips, multi-shot cuts and motion controlFrom 51 credits for a 5-second clip
• Vidu Q4 Preview APIUp to 4K clips with sound and spoken dialogueFrom 34 credits for a 5-second clip
• Wan 3.0 APIUp to 30-second clips with synced audioFrom 34 credits for a 5-second clip
• Kling O3 APIKling’s unified model: create, animate, reference and edit up to 4KFrom 51 credits for a 5-second clip
• LTX 2.5 APILong 20-second takes up to 4K with a soundtrackFrom 68 credits for a 5-second clip

More about MiniMax H3 Max

Try it

MiniMax H3 Max AI Video Generator onlineCreate videos with MiniMax H3 Max in PopNow

Docs

MiniMax H3 Max model ID and input schema in the API docsPopNow API quickstart: your first request in five minutesImage to video guide for the PopNow APIWebhooks: get MiniMax H3 Max results without polling