Try Kling Image V4 Long for Page Mockups from Kling AI →
Models
Agents
Workflows
Studio
PricingBlogDocs
ExploreDiscover models by categoryDiscoverModels and effects built by WiroBrowse All ModelsBrowse the complete catalogSee FavoritesSign in to view saved models
Generative Media AgentCreate and edit media by chattingWorkflow AgentBuild visual workflows with Agent
OverviewThe platform at a glanceLearnSkills, knowledge, guardrailsAnatomyWhat makes agents reasonBuild Your AgentPick connectors and skills, choose a plan, deploy
Pre-built AgentsBrowse the catalog
Agent Usecases
Ad Campaign ManagerApp Event ManagerApp Review RepliesBarber BookingCustomer Win-BackEcommerce ListingsRestaurant Reviews
Sign InStart Building
Explore models/
Video GenerationActive

MiniMax Hailuo 2.3 Fast

byminimax

The lighter variant of MiniMax Hailuo 2.3, for image-to-video only. You give it a first frame and an optional prompt, and it returns a 6 or 10 second clip with the image's proportions.

Image to VideoFast Inference
Model ID
hailuo-2.3-Fast
Provider
minimax
Updated
Oct 23, 2026, 21:08
6
Comments
Average rating : 4.64 (11 users)
Providerminimax
Modelhailuo-2.3-Fast
Image to VideoFast Inference
wiro playground—minimax/hailuo-2.3-Fast
1 / 1
Maximum 1 image allowed
Image 1

First frame image for video generation. The output video will have the same aspect ratio as this image.

The prompt to generate the video.

Running this model on Wiro costs $0.19 in total

Sample outputs
Updated Oct 23, 2026, 21:08

Overview

Hailuo 2.3 Fast is the lighter variant in MiniMax's Hailuo 2.3 release from October 2025, and it only does image-to-video. You supply the first frame and describe what should happen, and the clip it returns keeps that image's proportions. MiniMax offers it for batch creation, where many stills each need a moving version.

Resolution, duration, Prompt Optimizer and the bracketed camera commands work the same way as on the full model.

Choosing a variant: pick Fast when every clip starts from a picture you already have, such as a product catalog or a folder of key frames. Pick MiniMax Hailuo 2.3 when a shot has to start from a written idea alone, because Fast has no text-only mode.

What you can build

  • Moving versions of catalog photos, one clip per product image.
  • Several takes from one key frame, each with a different prompt or camera command, to compare before you commit to a shot.
  • Outdoor and sports footage from stills, like the surfer riding a breaking wave or the hiker climbing a snow track in this page's samples.
  • Ambient background clips, for example fish and a stingray drifting past aquarium glass while seated visitors watch from the foreground.

Inputs

  • Input Image (required): the opening frame, whose proportions carry over to the video. MiniMax takes JPG, PNG and WebP files below 20 MB. The shorter side must be over 300 pixels, and the aspect ratio has to fall from 2:5 to 5:2.
  • Prompt: optional, but it is where you say what moves and what the camera does. Up to 2,000 characters.
  • Prompt Optimizer: enabled unless you switch it off, in which case your prompt is used as written.
  • Resolution: 768P by default, or 1080P.
  • Duration: 6 seconds by default, or 10.

Bracketed commands such as [Tracking shot], [Push in] or [Static shot] steer the camera. Several inside one bracket run together (MiniMax recommends three at most), and separate brackets placed in order play one after the other.

Outputs

A clip of 6 or 10 seconds at 768P or 1080P that starts on your image.

Recommended settings

  • Describe motion in terms of what the image already shows: which subject moves, in which direction, and where the camera goes. The beach sample names each moving element in the frame, from the runner on the shoreline to the haze over the industrial stacks.
  • For a batch, keep one prompt template and change only the subject line, so the clips stay comparable.
  • Turn Prompt Optimizer off when the same instructions must apply to every image. MiniMax recommends this when precise control matters.
  • Crop each source image to the aspect ratio you plan to deliver before uploading it.

Limitations

  • An image is always required. A prompt alone will not start a run.
  • Ten-second clips come out at 768P. Choosing 1080P limits the clip to 6 seconds.
  • The image fixes the opening frame only. The end frame is left to the model.

Safety & compliance

Animate a photo of a real person only with their consent.

Example prompts

Great starting points for hailuo-2.3-Fast.

Static wide shot of the aquarium viewing area. The various fish and sharks swim gracefully and continuously in the deep blue water, with a large stingray gliding slowly past the glass. Ripples are visible on the water\'s surface above. The light shafts from above the water subtly undulate, creating a dynamic play of light and shadow in the tank. The silhouetted people in the foreground remain still, absorbed in the view. Submerged, serene, deep blue aesthetic.Video Generation
Tracking shot that follows the surfer as he rides the breaking wave toward the shore. The water sprays dynamically from the board, and the wave curls and crashes with realistic fluid motion. The golden hour light shimmers brightly on the water inside the tube of the wave. The surfer adjusts his weight and maintains balance. Cinematic slow motion, high-fidelity photorealism.Video Generation
Wide shot of the beach and power plant. The man continues his steady run along the shoreline, his footsteps leaving subtle indentations in the wet sand. The ocean waves gently roll in and recede, creating realistic foam and movement. The seagull in the sky gracefully soars and flaps its wings. Subtle steam or heat haze rises from the industrial stacks in the background. The light remains soft and hazy, as if late afternoon. Naturalistic, establishing shot.Video Generation
Tracking shot following the hiker as they steadily climb upward along the snow track. Their ski poles repeatedly sink and lift from the deep snow with realistic, labored effort. The camera slowly tilts up to keep the subject centered against the vast, clear blue sky. Extreme contrast, dramatic sunlight, realistic snow texture.Video Generation

API quick start

Run hailuo-2.3-Fast with a single API call.

POST https://api.wiro.ai/v1/Run/minimax/hailuo-2-3-fast
{
  "prompt": "Wide shot of the beach and power plant. T…",
  "inputImage": "https://your-cdn.com/input.png",
  "duration": 6,
  "promptOptimizer": "true"
}
curl
curl -X POST "https://api.wiro.ai/v1/Run/minimax/hailuo-2-3-fast" \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_WIRO_API_KEY" \
  --data-binary @- <<'JSON'
{
  "prompt": "Wide shot of the beach and power plant. T…",
  "inputImage": "https://your-cdn.com/input.png",
  "duration": 6,
  "promptOptimizer": "true"
}
JSON

Other models you might like

Explore similar video generation models.

View all
Image to Video

image-to-video-hailuo-01-subject

minimax

Image to Video

image-to-video-hailuo-01

minimax

Image to Video

image-to-video-hailuo-02

minimax

Image to Video

image-to-video-hailuo-01-director

minimax

  • image-to-video-hailuo-01-liveminimax
  • happyhorse 1.1 referencealibaba
  • Effectspixverse
  • image-to-video-gen4-turborunway
  • Transitionpixverse
  • image-to-video-v5pixverse
  • kling-v3-motion-controlklingai
  • image-to-video-v3.5pixverse

Discover, test, and run AI models, build workflows and agents with one unified API.

All systems operational
WiroModelsAgentsMCP ServerLLM GatewayBrowser ExtensionIntegrationsPricingAboutBlogCareersContactPartnerChangelogStatusFAQ
CategoriesText to ImageImage to ImageImage EditingImage to TextText to VideoImage to VideoVideo to VideoText to SpeechSpeech to TextSpeech to SpeechVoice CloneRealtime ConversationMusic GenerationLLMUtilities
ModelsNano Banana 2GPT Image 2.5Seedream V5 ProSeedance 2.5Veo 3.1Kling V3FLUX 3FLUX.2 ProWan 3.0 PrimeGrok Imagine 1.5
PartnersGoogleOpenAIByteDanceBlack Forest LabsKling AIQwenAlibabaxAIMiniMaxElevenLabs
Getting StartedWiro ModelsWiro EffectsIntroductionAuthenticationProjectsCode ExamplesWiro MCP ServerSelf-Hosted MCPn8n IntegrationLLMs.txt
API ReferenceModelsRun a ModelModel ParametersTasksLLM & Chat StreamingWebSocketRealtime VoiceFiles
© 2026 Wiro AI. All rights reserved.
PrivacyTermsData Deletion