Minimax Hailuo 2.3 I2V Pro
Playground
Try it on WaveSpeedAI!MiniMax Hailuo 2.3 Pro is an image-to-video model for ultra-clear 1080P output and physics-aware scenes with responsive rendering. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Features
Hailuo 2.3 Pro represents the pinnacle of MiniMax’s image-to-video generation technology. It turns static images into cinematic, full-HD 1080p videos with stunning realism, seamless motion, and physics-accurate animation. Built for creators, studios, and digital storytellers, it combines next-generation motion rendering with precise visual fidelity — all in just 5 seconds per clip.
🎬 Why It Looks Great
- Native 1080p Output – Generates crisp 1080p videos directly from the model (no upscaling), ensuring perfect clarity for big screens and professional use.
- Advanced Motion Dynamics – Accurately simulates camera motion, object interactions, lighting shifts, and depth transitions.
- Realistic Physics – Handles subtle natural effects — wind, fabric movement, reflections, and shadows — with organic consistency.
- Stable Composition – Keeps subject framing, proportions, and color tone stable across all frames.
- Cinematic Color and Lighting – Fine-tuned tone mapping for film-grade warmth and visual depth.
⚙️ Limits and Performance
- Input: one reference image (JPEG / PNG)
- Optional: text prompt to guide motion or camera style
- Duration: fixed at 5 seconds
- Resolution: 1080p
💰 Pricing
| Duration | Resolution | Cost per Job |
|---|---|---|
| 5 seconds | 1080p | $0.49 |
🚀 How to Use
- Upload a reference image to define your scene or subject.
- ( Optional ) Add a text prompt to describe motion, camera angle, or lighting. Example: “A woman walks through a sunlit market as the camera slowly pans right, dust particles drifting in the air.”
- Submit your job ( fixed 5 s duration ).
- Download your cinematic video.
💡 Pro Tips
- Use camera verbs like “pan,” “zoom,” “follow,” or “orbit” to define motion clearly.
- Mention lighting mood — soft afternoon light, golden hour, neon reflections — for enhanced aesthetic control.
- Include emotion or atmosphere in your prompt to guide tone (e.g., serene, tense, romantic).
- For stable results across iterations, reuse the same seed.
- Pair your video with music or voiceovers for a complete storytelling experience.
🧩 Why Creators Love It
> “Nothing else feels as cinematic straight out of the box — it just moves naturally.” > “The motion and lighting look like real camera work. Perfect for film concepts and game trailers.” > “The best image-to-video balance so far — clean details, zero flicker, amazing physics.”
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
set -euo pipefail
export WAVESPEED_API_KEY="your-api-key"
REQUEST_BODY=$(cat <<'JSON'
{
"image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
"enable_prompt_expansion": true
}
JSON
)
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/minimax/hailuo-2.3/i2v-pro" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
-H "Content-Type: application/json" \
-d "${REQUEST_BODY}")
TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
printf 'Submission response did not contain a prediction id
' >&2
exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi
# 2. Poll until the prediction finishes.
while true; do
RESPONSE=$(curl --silent --show-error --fail-with-body \
"${RESULT_URL}" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}")
RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')
case "${STATUS}" in
completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
created|processing) sleep 2 ;;
*) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
esac
doneParameters
Task Submission Parameters
Request Parameters
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| image | string | Yes | - | The model generates video with the picture passed in as the first frame.Base64 encoded strings in data:image/jpeg; base64,{data} format for incoming images, or URLs accessible via the public network. The uploaded image needs to meet the following conditions: Format is JPG/JPEG/PNG; The aspect ratio is greater than 2:5 and less than 5:2; Short side pixels greater than 300px; The image file size cannot exceed 20MB. | |
| prompt | string | No | - | The positive prompt for the generation. | |
| enable_prompt_expansion | boolean | No | true | - | The model automatically optimizes incoming prompts to enhance output quality. This also activates the safety checker, which ensures content safety by detecting and filtering potential risks. |
Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data.id | string | Unique identifier for the prediction, Task Id |
| data.model | string | Model ID used for the prediction |
| data.outputs | array | Output values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed) |
| data.urls | object | Object containing related API endpoints |
| data.urls.get | string | URL to retrieve the prediction result |
| data.status | string | Status of the task: created, processing, completed, or failed |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |
Result Request Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| id | string | Yes | - | Task ID |
Result Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data | object | The prediction data object containing all details |
| data.id | string | Unique identifier for the prediction |
| data.model | string | Model ID used for the prediction |
| data.outputs | array<string | object> | Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model. |
| data.urls | object | Object containing related API endpoints |
| data.urls.get | string | URL to poll for the prediction result |
| data.status | string | Status: created, processing, completed, or failed |
| data.created_at | string | ISO timestamp of when the request was created |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |