Black Forest Labs· released August 2026New
FLUX 3 Video
Up to 20 s video with native audio from Black Forest Labs
FLUX 3 Video from Black Forest Labs generates 5–20 second clips in 720p or 1080p with optional native audio, from text, a start image or a start and end frame. It also re-renders an existing clip from a text instruction while keeping its motion and timing (edits are rendered in 720p).
- Up to 20 seconds
- Start + end frame
- Optional native audio
- Prompt-based video edits
Capabilities
- Modes
- Text to video · Image to video · Video to video
- Aspect ratios
- 16:9 9:16 1:1 4:3 3:4 21:9
- Durations
- 5s 10s 15s 20s
- Resolutions
- 720p 1080p
- Inputs
- Up to 2 reference images
- Native audio
- ✓
Pricing
Credits per output.
| Setting | Quality | Credits |
|---|---|---|
| 1080p · Video to video | High | 70 |
| 720p · Video to video | Standard | 70 |
| 720p · 5s | High | 140 |
| 1080p · 5s | High | 240 |
| 720p · 10s | High | 280 |
| 720p · 15s | Ultra | 420 |
| 1080p · 10s | Ultra | 480 |
| 720p · 20s | Ultra | 560 |
| 1080p · 15s | Ultra | 720 |
| 1080p · 20s | Ultra | 950 |
Similar models
Popular
Veo 3.1
GoogleGoogle's flagship: cinematic video with synced sound
Text → VideoImage → VideoReferences → VideoNative audio
from 140 credits~100s
Popular
Kling 3.0
KuaishouCinematic 5–15 s video with optional native audio
Text → VideoImage → VideoNative audio
from 60 credits~300s
New
Seedance 2.5
ByteDanceByteDance's newest: up to 30 s in one take, with audio
Text → VideoImage → VideoReferences → VideoNative audio
from 140 credits~220s
New
MiniMax H3
MiniMaxOmni-modal 2K video with native stereo sound
Text → VideoImage → VideoReferences → VideoNative audio
from 95 credits~190s