HiAPI
flux-3-max 图生视频详情页 Hero 图:右侧玻璃胶片带与光轨延展,左侧柔光留白,冷调粉蓝与暖金交织,名字以发光立体字蚀刻在胶片带上。

Black Forest Labs

FLUX 3 Max

flux-3-max/image-to-video

A video generation model that supports text-to-video and image-to-video, focused on low-latency, fast turnaround; public benchmarks report roughly 11–12 seconds to generate a 10-second clip, making it well suited to creative iteration and short-video assets

Try flux-3-max/image-to-video

Input

Output

video
Fill in the parameters and run — results show up here.

Pricing

flux-3-max/image-to-video per second

Specifications Unit Price
720P $0.0608
1080P $0.1036

Overview

FLUX 3 Max image-to-video model: input a first-frame image and a prompt to generate a dynamic video clip that can include synchronized audio. It supports durations from 5 to 20 seconds, 720p and 1080p resolutions, 8 aspect ratios, and an optional last-frame image for first-to-last-frame transitions. It is suited for developers and content teams that need to quickly turn static images into usable video assets, callable via a unified API on HiAPI.ai.

Key Highlights

  • Dual first/last frame control — the first frame is required and the last frame is optional; by specifying the start and end images, you can constrain the shot's beginning and end and reduce composition drift in the generated result.
  • Synchronized audio generated with the video — audio generation is enabled by default, outputting video with synchronized sound and eliminating the later dubbing and track-alignment step.
  • Wide adjustable range for duration and aspect ratio — duration can be set from 5 to 20 seconds, and aspect ratios support 8 options from 21:9 to 9:16, or adapt to the input image.
  • Switchable reasoning level — reasoning_level offers low and high tiers; use the high tier for complex scenes to improve understanding, and the low tier for simple visuals to get lower latency.

Capabilities

First-frame-driven image-to-video: pass a publicly accessible HTTPS first-frame image (PNG, JPEG, WebP) and a prompt to generate a video that opens with that image, ideal for extending existing assets directly into dynamic clips.

Optional last-frame transition: after providing a last-frame image, the model generates a transition between the two given images, useful for shot continuity and transition asset creation with fixed start and end compositions.

Resolution and aspect ratio control: output can be 720p or 1080p, with aspect ratios covering 21:9, 16:9, 4:3, 1:1, 9:16, and more, or set to auto to follow the input image, making it easy to generate assets for different placements in one pass.

Set duration as needed: duration can be specified between 5 and 20 seconds, billed by output seconds, so short-clip validation and longer assets can be completed through the same API.

Synchronized audio toggle: generate_audio controls whether synchronized audio is generated with the visuals; disable it when you need silent footage for later scoring.

Use Cases

  • Design teams turning static key visuals into video — input a brand key visual and a prompt describing camera movement to produce a short video clip with audio for social media placements and landing page hero sections.
  • First/last-frame continuity in storyboard previews — given two images, the storyboard's first frame and last frame, generate the intermediate transition video as a shot draft for editing and client approval.
  • Vertical short-video content production — generate clips in 9:16 with shorter durations, directly usable for asset supplementation and re-editing on short-video platforms.
  • Rapid validation of creative concepts — generate multiple results from the same image with different aspect ratios, durations, and reasoning levels, quickly compare directions, and then finalize.

Limitations and Selection Advice

The first-frame image must be a publicly accessible HTTPS link, and prompts are capped at 6000 characters; trim them yourself if they exceed this. Long durations and high resolutions increase generation time, so batch tasks should allow for a waiting window. This model only does image-to-video; when controllable reference images are lacking, a text-to-video model is more suitable. If you need a fixed rhythm across multiple keyframes, consider the same family's keyframes-to-video route. For complex scenes, use the high reasoning level and manually verify the output.

  • New model flux-3-max/text-to-video video generation model now available
  • Price update flux-3-max/text-to-video Pricing Adjustment
  • Price update flux-3-max/image-to-video price update: new pricing at $0.0608

Frequently asked questions

View all FAQs →
  • What inputs does FLUX 3 Max Image to Video need?

    prompt and image_url (the first frame, PNG/JPEG/WebP) are required. Optional: end_image_url as the last frame, plus duration, resolution, aspect_ratio, generate_audio and reasoning_level.

  • Can I set the video length?

    Yes. duration accepts whole seconds from 5 to 20, 5 by default. Image to video shares the same per-second pricing as text to video, and a 5-second check shows whether the source stays stable.

  • What decides the aspect ratio?

    aspect_ratio defaults to auto, which follows the first frame's ratio. You can also set 21:9, 2:1, 16:9, 4:3, 1:1, 3:4 or 9:16. If a manual ratio differs from the image, the frame may be cropped.

  • How do first and last frames work?

    Pass the first frame in image_url and the last in end_image_url; the video moves from one to the other. The prompt only needs to describe the change in between, and the closer the two frames match in framing and light, the smoother the transition.

Related Models

View all models →

Same Type