Hyference

Hyference / Models

AI models

Original model names, real limits and the price you pay. 11 video models and 6 image models are live in the Studio today; the models marked coming soon need inputs we don't support yet.

FASTLive · Google

Gemini Omni Flash 1.1

by Google

Fast drafts for feeds

Powered by Google

Version 1.1 Flash

Fast video drafts for social feeds, UGC concepts, and product shots. The lowest-cost option in the Studio for trying different creative directions.

Rate / second$0.20
Longest clip (10s)$2.00
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
3, 4, 5, 6, 7, 8, 9, 10s
Resolution
720p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
NEWLive · Google

Veo 3.1 Fast

by Google

Fluid, physics-aware motion

Powered by Google

Version 3.1 Fast

A motion-focused option for flowing movement, runway scenes, and camera tracking. A balanced choice between quick drafts and premium cinematic shots.

Rate / second$0.20
Longest clip (8s)$1.60
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
4, 6, 8s
Resolution
720p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
TOPLive · Google

Veo 3.1

by Google

Cinematic, film-grade output

Powered by Google

Version 3.1

The Studio’s premium cinematic option for expressive lighting, composed frames, and film-inspired sequences. Suited to polished brand films and cinematic concepts.

Rate / second$0.80
Longest clip (8s)$6.40
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
4, 6, 8s
Resolution
1080p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
TOPLive · Kuaishou

Kling 3.0 Pro

by Kuaishou

Cinematic, fluid motion

Powered by Kuaishou

Version 3.0 Pro

Top-tier cinematic motion and detail. Audio off.

Rate / second$0.22
Longest clip (10s)$2.24
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
3, 4, 5, 6, 7, 8, 9, 10s
Resolution
1080p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
NEWLive · Alibaba

Wan 3.0

by Alibaba

Smooth motion, strong scenes

Powered by Alibaba

Version 3.0

Latest Wan generation with smooth motion and coherent scenes, 720p.

Rate / second$0.20
Longest clip (10s)$2.00
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
3, 4, 5, 6, 7, 8, 9, 10s
Resolution
720p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
NEWLive · MiniMax

MiniMax H3

by MiniMax

Frontier model, 768p

Powered by MiniMax

Version H3

MiniMax's frontier video model at native 768p.

Rate / second$0.12
Longest clip (10s)$1.20
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
5, 6, 7, 8, 9, 10s
Resolution
768p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
Live · MiniMax

Hailuo 02 Pro

by MiniMax

1080p, strong physics

Powered by MiniMax

Version 02 Pro

Realistic physics and camera control at 1080p. Fixed 6s clips, landscape.

Rate / second$0.16
Longest clip (6s)$0.96
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
6s
Resolution
1080p
Aspect ratios
16:9
Audio
Not available
Select model
FASTLive · MiniMax

Hailuo 2.3

by MiniMax

Fast, high-dynamic drafts

Powered by MiniMax

Version 2.3 Standard

Affordable high-dynamic video at 768p, 6s or 10s, landscape.

Rate / second$0.09
Longest clip (10s)$0.94
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
6, 10s
Resolution
768p
Aspect ratios
16:9
Audio
Not available
Select model
Live · Lightricks

LTX-2.3

by Lightricks

Open model, sharp 1080p

Powered by Lightricks

Version 2.3

Lightricks' open video model at 1080p. Audio off.

Rate / second$0.16
Longest clip (10s)$1.60
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
6, 8, 10s
Resolution
1080p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
Live · Shengshu

Vidu Q3

by Shengshu

Expressive motion, 720p

Powered by Shengshu

Version Q3 Pro

Vidu's latest Q3 model with expressive motion at 720p.

Rate / second$0.31
Longest clip (10s)$3.08
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
3, 4, 5, 6, 7, 8, 9, 10s
Resolution
720p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model
FASTLive · PixVerse

PixVerse v5.5

by PixVerse

Cheapest social clips

Powered by PixVerse

Version 5.5

Budget-friendly clips for social feeds at 720p.

Rate / second$0.10
Longest clip (10s)$1.00
Modes
Text-to-video, Image-to-video
Inputs
Prompt, reference image, first & last frames
Durations
5, 8, 10s
Resolution
720p
Aspect ratios
16:9, 9:16
Audio
Not available
Select model

More video models on Hyference

Coming soon

Luma Ray 3.2

by Luma

Restyle an existing video

$0.29/s

540p video-to-video; needs a source video

Image models

GPT Image 2.5 Sunburst

by OpenAI

Versatile, follows prompts closely

$0.25/image

Accepts reference and character photos

Select model

Nano Banana Pro

TOP

by Google

Best identity and edit consistency

$0.30/image

Accepts reference and character photos

Select model

FLUX.2

NEW

by Black Forest Labs

Sharp photoreal, multi-reference edits

$0.08/image

Accepts reference and character photos

Select model

FLUX 1.1 Pro Ultra

by Black Forest Labs

Photoreal, up to 2K

$0.12/image

Prompt only

Select model

Ideogram V3

by Ideogram

Posters and readable text

$0.06/image

Prompt only

Select model

FLUX.1 Schnell

FAST

by Black Forest Labs

Sub-second drafts

$0.02/image

Prompt only

Select model

Each Studio model routes to a distinct real video engine with its own duration and resolution limits, shown above. Optional settings (camera, motion intensity, frames) don't change the price — only the duration does.