NVIDIA SANA-Sprint

Unleash creativity with NVIDIA SANA-Sprint, an ultra-efficient one-step diffusion model. Ready to experience the power of AI? Start your journey here!

Platform: Replicate

Text-to-Image DiffusionFast Image SynthesisConsistency DistillationFew-Step Generation

289.6k runs

H100

License Check Required

🚀Function Overview

An ultra-efficient diffusion model that generates images from text prompts in 1-4 steps using continuous-time consistency distillation and ControlNet integration.

Key Features

Generates 1024px images in 0.1s on H100 hardware
Unified step-adaptive architecture (1-4 steps)
Continuous-time consistency distillation without retraining
ControlNet support for interactive generation
State-of-the-art quality in few-step generation

Use Cases

•Artwork and design asset creation
•Educational/creative content tools
•Research on efficient generative models
•Real-time image synthesis applications

⚙️Input Parameters

prompt

string

Input prompt

width

integer

Width of output image

height

integer

Height of output image

inference_steps

integer

Number of sampling steps

intermediate_timesteps

number

Intermediate timestep value (only used when inference_steps=2, recommended values: 1.0-1.4)

guidance_scale

number

CFG guidance scale

seed

integer

Seed value. Set to a value less than 0 to randomize the seed

output_format

string

Format of the output images

output_quality

integer

Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs

💡Usage Examples

Example 1

Input Parameters

{
  "seed": -1,
  "width": 1024,
  "height": 1024,
  "prompt": "a photograph of an astronaut riding a horse",
  "guidance_scale": 4.5,
  "inference_steps": 2
}

Output Results

https://replicate.delivery/xezq/8yRdmvrc3IKbIB35efMDGE10ELPuiQc2sw3iHfweIAXEt54RB/output.png

Quick Actions

Use NowView Documentation

Technical Specifications

Hardware Type: H100
Run Count: 289.6k
Commercial Use: Unknown/Restricted
Platform: Replicate

Related Keywords

One-Step DiffusionFast Image SynthesisContinuous-Time Consistency DistillationControlNet SupportFew-Step GenerationArtwork and Design Asset CreationReal-Time Image Synthesis

Related Models

Mindy Generator

A model for generating images using text prompts with support for image-to-image translation, inpainting, and LoRA integration.

HDPE Plastic Container Generator

a translucent 5 litres plastic container made from HDPE, for storage and transport of liquid, flavored drink concentrate.

Google Imagen 4 Fast

Use this fast version of Imagen 4 when speed and cost are more important than quality