Nitro-E 1024px - Diffusers Integration

This is the Nitro-E 1024px text-to-image diffusion model in diffusers format.

Model Description

Nitro-E is a family of text-to-image diffusion models focused on highly efficient training. With just 304M parameters, Nitro-E is designed to be resource-friendly for both training and inference.

Key Features:

304M parameters
Efficient training: 1.5 days on 8x AMD Instinct MI300X GPUs
High throughput: Optimized samples/second on single MI300X
Consumer GPU support: Fast per 1024px image on Strix Halo iGPU

Model Variant

This is the 1024px variant, optimized for generating 1024x1024 images.

Note: This variant uses standard attention (no ASA subsampling).

Original Model

This model is based on amd/Nitro-E and has been converted to the diffusers format for easier integration and use.

Usage

import torch
from diffusers import NitroEPipeline

# Load pipeline
pipe = NitroEPipeline.from_pretrained("blanchon/nitro_e_1024", torch_dtype=torch.bfloat16)
pipe = pipe.to("cuda")

# Generate 1024x1024 image
prompt = "A hot air balloon in the shape of a heart grand canyon"
image = pipe(
    prompt=prompt,
    width=1024,
    height=1024,
    num_inference_steps=20,
    guidance_scale=4.5,
).images[0]

image.save("output.png")

Technical Details

Architecture

Type: E-MMDiT (Efficient Multi-scale Masked Diffusion Transformer)
Attention: Standard attention
Text Encoder: Llama-3.2-1B
VAE: DC-AE-f32c32 from MIT-Han-Lab
Scheduler: Flow Matching with Euler Discrete Scheduler
Sample Size: 32 (latent space)

Training

Dataset: ~25M images (real + synthetic)
Duration: 1.5 days on 8x AMD Instinct MI300X GPUs
Training Details: See Nitro-E Technical Report

Citation

If you use this model, please cite:

@article{nitro-e-2025,
  title={Nitro-E: Efficient Training of Diffusion Models},
  author={AMD AI Group},
  journal={arXiv preprint arXiv:2510.27135},
  year={2025}
}

License

Licensed under the MIT License. See the LICENSE for details.

Related Projects

Nitro-T: Efficient Training of diffusion models
Nitro-1: One-step distillation of diffusion models
Original Nitro-E Repository
AMD Nitro-E on HuggingFace