Models

The models behind every Tonta piece

Tonta routes each shot to the model that fits your brief and your “optimise for” setting — speed, quality or cost. Browse what's under the hood below, or leave the choice to us.

In motion

Every model, in motion

Real Tonta renders, one per model below. Hover a frame to play it, or open the prompt that made it.

A drone-style push through a foggy pine forest at dawn, demonstrating dynamic camera motion.
Kling 2.5 Turbo (Pro)

Generated by Tonta with Kling 2.5 Turbo (Pro).

Prompt

A sweeping cinematic drone-style push through a foggy pine forest at dawn, dynamic camera motion, mist swirling with realistic physics, shallow depth of field, no people, no text, no logos.

A paper airplane gliding through a softly lit workshop, demonstrating naturalistic motion.
MiniMax Hailuo 02 (Standard)

Generated by Tonta with MiniMax Hailuo 02 (Standard).

Prompt

A paper airplane gliding and looping through a softly lit workshop, gentle realistic physics, warm rim light, no people, no text, no logos.

An old brass clock mechanism turning in extreme macro, demonstrating intricate detail.
Wan 2.5 (Preview)

Generated by Tonta with Wan 2.5 (Preview).

Prompt

An old brass clock mechanism turning in extreme macro, intricate moving gears, cinematic lighting, shallow depth of field, no people, no text, no logos.

Rain droplets sliding down a window with glowing city bokeh beyond.
LTX Video 13B (Distilled)

Generated by Tonta with LTX Video 13B (Distilled).

Prompt

Rain droplets sliding down a dark window with soft glowing city bokeh beyond, slow motion, moody blue-grey grade, no people, no text, no logos.

A single feather falling through a shaft of window light, demonstrating photorealistic physics.
Veo 3.1 (Lite)

Generated by Tonta with Veo 3.1 (Lite).

Prompt

A single feather falling in slow motion through a shaft of window light in a quiet room, photorealistic, natural physics, no people, no text, no logos.

A drop of ink blooming and unfurling in still water, extreme macro.
Seedance 2.0

Generated by Tonta with Seedance 2.0.

Prompt

A single drop of ink blooming and unfurling in perfectly still water, extreme macro, intricate swirling detail, no people, no text, no logos.

Seedance 2.5 reference-to-video: two unlabeled skincare products from separate reference stills, arranged together in a new scene, holding both products consistent across references.
Seedance 2.5 · product consistency across references

Generated by Tonta with Seedance 2.5 · product consistency across references.

Prompt

@Image1 is a set of unlabeled amber dropper bottles; @Image2 is an unlabeled open cream jar. A hand arranges both together on a sunlit bathroom shelf, slow push-in, hands only, no face.

Seedance 2.5 text-to-video: a two-shot cinematic sequence of a Nigerian music producer in a home studio, with camera movement and native ambient audio.
Seedance 2.5 · multi-shot with sound

Generated by Tonta with Seedance 2.5 · multi-shot with sound.

Prompt

A multi-shot cinematic sequence: a Nigerian music producer leaning over a mixing desk with a slow push-in, cutting to him leaning back smiling as string lights blink behind him on a slow pull-back, with ambient room tone and a soft synth hum.

A colourful canvas sneaker product photo, the input a 3D model was generated from.
Made with Hunyuan3D
Hunyuan3D

A 3D model generated from this product photo, with Hunyuan3D.

LTX 2.3Every plan
Signature 1080
Image to video · video
  • 1080P
  • 6–10s clips
  • Holds a subject from one reference image
  • 16:9, 9:16
LTX 2.3Every plan
Signature 1080 from a brief
Text to video · video
  • 1080P
  • 6–10s clips
  • 16:9, 9:16
Kling 3.0Every plan
Quick Cut
Text to video · video
  • 3–15s clips
  • 16:9, 9:16, 1:1
LTX 2.3Every plan
Everyday 1080
Image to video · video
  • 1080P
  • 6–10s clips
  • Holds a subject from one reference image
  • 16:9, 9:16
LTX 2.3Every plan
Everyday 1080 from a brief
Text to video · video
  • 1080P
  • 6–10s clips
  • 16:9, 9:16
Seedance 2.5Every plan
Premium Motion
Image to video · video
  • 480P / 720P
  • 4–30s clips
  • Holds a subject across up to 2 reference images
Seedance 2.5Every plan
Premium Motion from references
Reference to video · video
  • 480P / 720P
  • 4–30s clips
  • Holds a subject across up to 30 reference images
  • 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
Z-Image TurboEvery plan
Still
Image · image
  • 1:1, 4:3, 3:4, 16:9, 9:16
Clarity UpscalerEvery plan
Sharpen 2K
Image · image
  • Holds a subject from one reference image
Veo 3.1Every plan
Cinematic Motion
Image to video · video
  • 720P / 1080P
  • 4–8s clips
  • Holds a subject from one reference image
  • 16:9, 9:16
Veo 3.1Every plan
Cinematic Motion from a brief
Text to video · video
  • 720P / 1080P
  • 4–8s clips
  • 16:9, 9:16
Veo 3.1 FastEvery plan
Cinematic Motion, faster
Image to video · video
  • 720P / 1080P
  • 4–8s clips
  • Holds a subject from one reference image
  • 16:9, 9:16
Kling 2.5 Turbo ProEvery plan
Fluid Motion Pro
Image to video · video
  • 5–10s clips
  • Holds a subject across up to 2 reference images
Kling 3.0 Turbo ProEvery plan
Fluid Motion 3.0
Image to video · video
  • 3–15s clips
  • Holds a subject from one reference image
Kling 3.0 Turbo StandardEvery plan
Fluid Motion 3.0 from a brief
Text to video · video
  • 3–15s clips
  • 16:9, 9:16, 1:1
Wan 3.0 PrimeEvery plan
Everyday Prime
Image to video · video
  • 480P / 720P / 1080P
  • 2–30s clips
  • Holds a subject across up to 2 reference images
  • 16:9, 4:3, 1:1, 3:4, 9:16
Wan 3.0 PrimeEvery plan
Everyday Prime from a brief
Text to video · video
  • 480P / 720P / 1080P
  • 2–30s clips
  • 16:9, 4:3, 1:1, 3:4, 9:16
Hailuo 02 ProEvery plan
Motion Pro
Image to video · video
  • 6s clips
  • Holds a subject across up to 2 reference images
Hailuo 02 StandardEvery plan
Motion Standard from a brief
Text to video · video
  • 6–10s clips
PixVerse V5.5Every plan
Short Beat 1080 V5
Image to video · video
  • 720P / 1080P
  • 5–10s clips
  • Holds a subject from one reference image
PixVerse V5.6Every plan
Short Beat 1080 V5 from a brief
Text to video · video
  • 720P / 1080P
  • 5–10s clips
  • 16:9, 4:3, 1:1, 3:4, 9:16
FLUX.1 Kontext ProEvery plan
Still: Match to Brief
Image · image
  • 21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21
FLUX1.1 Pro UltraEvery plan
Still Ultra
Image · image
  • 21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21
Nano BananaEvery plan
Still: Quick Edit
Image · image
  • 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16
Nano Banana ProEvery plan
Still: Detail Pass
Image · image
  • 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16
Seedream 5Every plan
Still: Flash 2K
Image · image
  • 1:1, 4:3, 3:4, 16:9, 9:16

Behind every piece

What else goes into a piece

A reel is more than one generated shot. These are the models Tonta calls for the parts around it — not something you pick, just part of how a piece gets made.

Narration
MiniMax Speech 2.8 HD
Turns a written script into narration audio.
Music
ElevenLabs Music
Composes a licensed music bed to sit under a piece.
Sound effects
ElevenLabs Sound Effects
Generates short sound effects to layer into a piece.
Lip sync
PixVerse Lip Sync
Re-times a dubbed voice track to the speaker's mouth movements when a piece is localised.
Captions
ElevenLabs Scribe v2
Transcribes spoken audio into timed, speaker-labelled captions.