Video

Flux 3 API

Text, images, or a clip in, 20 seconds of video out.

Black Forest Labs' first video model through one unified API. Up to 20 seconds of 720p or 1080p video with synchronized audio, from a prompt, up to 10 images, or a clip you want to continue. A single REST endpoint, async jobs, and webhooks.

Get API key Read the docs

20s
Max clip length: 1080p
Max resolution: 10
Storyboard images: 8 credits/s
Draft from

Video · example

Playground

Try Flux 3 right here.

model: flux-3

Prompt

Aspect ratio Duration Resolution

Generate audio Synchronized ambient sound, speech, and effects. On by default.

Media inputs

Images Start video

Advanced settings

Draft preview Fast, low-cost 720p preview for iterating on a prompt.

Estimated cost 64–528 credits · $0.64–$5.28 Generate in Studio

Output Example

Vertical shot: a night-shift attendant in a navy parka stands under a fluorescent gas-station canopy in heavy rain, steam from a paper coffee cup, radio weather report and rain on metal, slow push-in, photoreal.

Capabilities

What's new in Flux 3.

Pay per second of video

Billing scales with the clip you ask for: 21 credits ($0.21) per second at 720p and 36 credits ($0.36) per second at 1080p, audio included. A 5-second 720p clip is 105 credits ($1.05); a 10-second 1080p clip is 360 credits ($3.60).

Audio in the same pass

Ambient sound, speech, and effects are generated with the video and synchronized to it, at no extra cost. Turn it off with a single flag when you want a silent clip.

Up to 20 seconds in one generation

Ask for any whole number of seconds from 5 to 20, or let the model choose a length that fits the content.

Storyboard from up to 10 images

One image opens the clip pixel for pixel, two set the first and last frame, and three or more become keyframes spread evenly across the timeline.

Continue a shot, then chain it

Pass a finished mp4 as start_video and Flux 3 picks up from its final frames, so you can build sequences well past the 20-second per-generation limit.

Cheap drafts while you iterate

Draft mode returns a fast 720p preview at 8 credits ($0.08) per second, roughly a third of the full-quality rate, so you can test prompts before paying for final quality.

Showcase

See what Flux 3 can create.

Real outputs generated with Flux 3 on Apiframe, each with the prompt behind it.

Try Flux 3

A fluorescent-lit gas-station shop at night, rain streaking the glass, a coffee machine hissing steam, radio chatter and rain on the roof, slow push-in, photoreal.

Close-up through a rain-streaked windshield at night, wipers sweeping, dashboard glow, a paper map on the passenger seat, rain and wiper rhythm on the soundtrack, photoreal.

Ultra-wide aerial of a lone gas station on an empty highway at night, rain, one rust sedan at the pumps, fluorescent canopy glowing in the dark, distant thunder and rain, slow crane up, photoreal.

Multi-shot night sequence at an isolated rainy gas station. Wide: rain hammering the fluorescent canopy, a rust sedan idling at pump three. Cut to a close-up of the driver, wet windshield, dashboard glow, he mutters almost home. Rain, engine idle, distant thunder, wipers. Cinematic, photoreal.

Pricing

Simple per-second pricing.

8–66 credits per second of output ≈ $0.08–$0.66

The rate depends on the resolution and options you pick.

A 12s clip runs 96–792 credits (≈ $0.96–$7.92).

You only pay for successful generations. Failed jobs are refunded automatically.
One credit balance across every model on Apiframe, no per-model plans.
Start with free credits. No subscription required.
Volume discounts on larger plans.

See plans and credit packs API pricing Flux 3 credit cost and rates

Specs

At a glance.

Provider Black Forest Labs

Model ID flux-3

Aspect ratios auto, 16:9, 21:9, 2:1, 4:3, 1:1, 3:4, 9:16

Resolutions 720p, 1080p

Durations 5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s, 16s, 17s, 18s, 19s, 20s

Image input Supported

Audio Supported

Avg. time ~165s

Comparison

Flux 3 vs Seedance 2.5 vs Veo 3.1

Flux 3 against the other long-form video models on Apiframe. One parameter swap on the same key.

Feature	Flux 3	Seedance 2.5	Veo 3.1
Max clip length	20s (chainable via start_video)	30s in a single pass	8s
Max resolution	1080p	720p	1080p
Image inputs	Up to 10 storyboard keyframes	Up to 30 reference images	Start/end frames + 3 references
Native audio	Co-generated	Co-generated with lip-sync	Co-generated
Price from	8 credits/s draft, 21 credits/s final (≈$0.21)	13 credits/s (≈$0.13)	100 credits/clip (≈$1.00)
Best for	Storyboarded long takes and chained shots	Single-pass narrative with references	Filmic single shots

Specs reflect what each model supports on Apiframe today, from the live catalog.

Quickstart

How to call the Flux 3 API.

Send one POST /v2/videos/generate request with your API key. The call returns a jobId you can poll, or pass a webhook_url and the result is pushed to you the moment it's ready.

Swap flux-3 for any other model and nothing else changes.

Get API key Read the docs

generate.js

const response = await fetch("https://api.apiframe.ai/v2/videos/generate", {
  method: "POST",
  headers: {
    "X-API-Key": "afk_your_api_key_here",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "prompt": "a cinematic sunrise over a futuristic cityscape, smooth camera push-in",
    "model": "flux-3",
    "flux3Params": {
      "images": [
        "https://example.com/input.jpg"
      ],
      "start_video": "https://example.com/input.jpg",
      "resolution": "720p",
      "generate_audio": true
    }
  }),
});
const { jobId } = await response.json();
console.log(jobId);

import requests

response = requests.post(
    "https://api.apiframe.ai/v2/videos/generate",
    headers={
        "X-API-Key": "afk_your_api_key_here",
        "Content-Type": "application/json",
    },
    json={
        "prompt": "a cinematic sunrise over a futuristic cityscape, smooth camera push-in",
        "model": "flux-3",
        "flux3Params": {
            "images": [
                "https://example.com/input.jpg"
            ],
            "start_video": "https://example.com/input.jpg",
            "resolution": "720p",
            "generate_audio": True
        }
    },
)
print(response.json())  # { "jobId": "...", "status": "QUEUED" }

curl -X POST https://api.apiframe.ai/v2/videos/generate \
  -H "X-API-Key: afk_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
        "prompt": "a cinematic sunrise over a futuristic cityscape, smooth camera push-in",
        "model": "flux-3",
        "flux3Params": {
            "images": [
                "https://example.com/input.jpg"
            ],
            "start_video": "https://example.com/input.jpg",
            "resolution": "720p",
            "generate_audio": true
        }
    }'

let job;
do {
  await new Promise((r) => setTimeout(r, 2000));
  job = await fetch("https://api.apiframe.ai/v2/jobs/JOB_ID", {
    headers: { "X-API-Key": "afk_your_api_key_here" },
  }).then((r) => r.json());
} while (job.status !== "COMPLETED" && job.status !== "FAILED");
console.log(job.result);

import requests, time

while True:
    job = requests.get(
        "https://api.apiframe.ai/v2/jobs/JOB_ID",
        headers={"X-API-Key": "afk_your_api_key_here"},
    ).json()
    if job["status"] in ("COMPLETED", "FAILED"):
        break
    time.sleep(2)
print(job["result"])

curl https://api.apiframe.ai/v2/jobs/JOB_ID \
  -H "X-API-Key: afk_your_api_key_here"

Input schema

Every field Flux 3 accepts, with types and defaults. The full reference lives in the docs.

Field	Type	Description
prompt required	string	Text description of what to generate.
model required	string	The model identifier for this endpoint. Default:"flux-3" · "flux-3"
flux3Params.images	string[] (URLs)	None = text-to-video, one opens the clip, two start and end it, three or more become a storyboard (pick a duration).
flux3Params.start_video	string (URL)	Continue a clip from its final frames. Cannot be combined with images.
flux3Params.resolution	string	Resolution Default:"720p" · "720p", "1080p"
flux3Params.generate_audio	boolean	Synchronized ambient sound, speech, and effects. On by default. Default:true
flux3Params.draft	boolean	Fast, low-cost 720p preview for iterating on a prompt. Default:false

Flux versions

Every Flux video model on Apiframe. Switch with one parameter.

Flux 3 Flagship This page

Use cases

What teams build with Flux 3.

Storyboard-to-video

Up to 10 images become keyframes spread across the clip, sketch the beats as stills, and Flux 3 films between them.

Long-take sequences

Up to 20 seconds per generation, and start_video continuation picks up from the final frames of an existing mp4 to chain shots indefinitely.

Cheap drafts, final quality later

Draft mode renders fast 720p previews at 8 credits per second, roughly a third of final cost, so you burn credits only on the take that works.

Sound in the same pass

Ambient noise, speech, and effects are co-generated and synchronized at no extra cost, with a single flag to go silent.

Reviews

9,800+ developers. One API.

Real Apiframe reviews from Trustpilot and G2.

Seamless experience, excellent support

Just created 6 videos and the experience has been seamless. Wanted to write a review for appreciation — the customer support has also been excellent. They helped me in some process I was stuck in.

Vaibhav Sharma July 2026 Trustpilot

Stable, easy image & video generation

I like that I can generate images and videos without complicated setup. It works stable and is easy to use even if you are not a big developer. It saves me time and makes the whole process much simpler.

Anakin S. July 2026 G2

Honestly the best AI API

They got every engine possible and it's great to have them all in one place.

Daniel Zborowski July 2026 Trustpilot

Flux 3 questions

What teams ask before shipping with Flux 3. Everything else lives in the docs.

How long can a Flux 3 video be?

Any whole number of seconds from 5 to 20 in a single generation. Omit the duration and the model picks a length that fits your prompt, that request is billed at the 20-second maximum, so pass an explicit duration for exact pricing. To go longer, pass a finished clip as start_video and chain generations together.

What does Flux 3 cost on Apiframe?

It is billed per second of video: 21 credits ($0.21) per second at 720p and 36 credits ($0.36) per second at 1080p, with audio included. Draft previews are 8 credits ($0.08) per second. Continuing an existing video costs 51 credits ($0.51) per second at 720p and 66 credits ($0.66) per second at 1080p. Omitting the duration is billed at the 20-second maximum.

Does it generate audio?

Yes. Synchronized ambient sound, speech, and effects are produced in the same pass as the video and are on by default, at no extra charge. Set generate_audio to false for a silent clip.

How do I control the first and last frame?

Pass images in flux3Params. One image opens the clip as its first frame, shown pixel for pixel. Two images set the start and the end. Three or more (up to 10) act as a storyboard, evenly spaced across the clip, that mode needs an explicit duration.

Can I extend a video I already generated?

Yes. Pass its URL as flux3Params.start_video and Flux 3 continues from the final frames. The input must be an mp4 of at most 50MB and 15 seconds, and it cannot be combined with images.

How is Flux 3 different from Flux 2?

Flux 2 is an image model. Flux 3 is Black Forest Labs’ multimodal model built on their Self-Flow architecture, trained across images, video, and audio, and it outputs video with sound.

What resolutions and aspect ratios does it support?

720p or 1080p, in 21:9, 2:1, 16:9, 4:3, 1:1, 3:4, or 9:16. Leave the aspect ratio on auto and the model picks one from your prompt and inputs.