FLUX 3

Multi-shot video generation with native audio from text, images, or keyframes

About model

FLUX 3 is Black Forest Labs' video generation model, built on a unified multimodal architecture that generates video and synchronized audio together. It produces clips of up to 20 seconds with multiple scenes and camera angles in a single take, starting from text, an image, or ordered keyframes. Audio is optional and native: multilingual speech with strong lipsync and accurate accents, plus sound effects and ambience generated with the frames. The model spans styles well beyond cinematic output, renders text and typography that feels native to the scene, and supports video continuation, extending an existing clip from its final frame with momentum, framing, and scene logic carried forward. Built on Self-Flow, Black Forest Labs' approach to aligning multimodal generation and understanding within one architecture. Available on Together AI.

‍

Max Clip Length

20s

Multiple scenes and camera angles in a single generation

Audio Generation

Native

Multilingual speech with strong lipsync, plus effects and ambience, generated with the frames

Frame-Level Control

Keyframes

Start, end, and ordered in-between frames, plus continuation from a clip's final frame

Model key capabilities

Multi-Shot Generation: Multiple scenes and camera angles in one clip, with agentic chaining into longer cohesive stories
Native Audio: Optional multilingual speech with strong lipsync and accents, plus sound effects and ambience generated with the frames
Keyframe & Continuation Control: Direct generations with start, end, and ordered in-between keyframes, or extend an existing clip from its final frame
Production-Ready Infrastructure: 99.9% SLA, available on Together AI serverless infrastructure

API usage

cURL
Python
Typescript

Endpoint:

black-forest-labs/FLUX-3

curl --request POST \
  --url https://api.together.xyz/v2/videos \
  --header "Authorization: Bearer $TOGETHER_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "black-forest-labs/FLUX-3",
    "prompt": "some penguins building a snowman"
  }'

from together import Together

client = Together()

# Create a video generation job
job = client.videos.create(
    prompt="A serene sunset over the ocean with gentle waves",
    model="black-forest-labs/FLUX-3"
)

import Together from "together-ai";

const together = new Together();

async function main() {
  // Create a video generation job
  const job = await together.videos.create({
    prompt: "A serene sunset over the ocean with gentle waves",
    model: "black-forest-labs/FLUX-3"
  });

Model card
Architecture Overview:
• Unified multimodal architecture generating video and synchronized audio together, part of Black Forest Labs' FLUX 3 family spanning video, audio, and image generation
• Built on Self-Flow, Black Forest Labs' method for aligning multimodal generation and understanding within the same underlying architecture
• Generates clips of up to 20 seconds with multi-scene structure in a single pass

Training Methodology:
• Black Forest Labs reports Self-Flow achieves lower generation error across modalities than standard flow matching, and higher success rates on manipulation tasks through fine-tuning

Performance Characteristics:
• Handles simple and complex prompts across a broad stylistic range beyond conventional cinematic output
• Accurate multilingual speech with strong lipsync and accents in generated dialogue
• Text and typography rendered natively within scenes
• Video continuation carries momentum, framing, and scene logic forward from an existing clip's final frame
‍
Prompting
Together AI API Access:
• Access FLUX 3 via Together AI APIs using the endpoint black-forest-labs/FLUX-3
• Authenticate using your Together AI API key in request headers
• Create a video request with a text prompt, optionally providing an image or ordered keyframes, then poll until the video completes
• Extend an existing clip by continuing from its final frame
• Available on Together AI serverless infrastructure
‍
Applications & use cases
Filmmaking & Storyboarding:
• Generate multi-scene sequences with consistent direction in a single request
• Previsualize shots with keyframe control over start, end, and in-between frames
• Chain clips into longer cohesive stories through continuation from final frames

Advertising & Brand Content:
• Produce product and brand video with native typography rendered into the scene
• Generate campaign variations across styles from the same creative brief
• Add multilingual voiceover and dialogue with lipsync in the same generation

Explainers & Social Content:
• Create explainer videos and social clips with speech, effects, and ambience in one pass
• Turn a single image into a moving, sounding clip for feeds and stories
• Batch-generate localized variants with multilingual dialogue through the Together API
‍

Related models

Model specifications

Model data

Model provider
Black Forest Labs
Type
Video
Resolution/Duration
Up to 20s
Deployment
Serverless
Endpoint
black-forest-labs/FLUX-3
Price
$0.17 / video estimated per video
Input modalities
Text
Image
Video
Output modalities
Video

Category
Video

Run in Playground

Quickstart docs

Deploy model

FLUX 3

About model

API usage

Model card

Prompting

Applications & use cases