LogoWaveSpeedAI

wavespeed-ai/hidream-i1-full

HiDream-I1 is a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds.

text-to-image

width
height
If set to true, the output base64 will be enabled.
The safety checker cannot be disabled on the playground. This property is only available through the API.
If set to true, the safety checker will be enabled.

Idle

https://d2g64w682n9w0w.cloudfront.net/media/images/1744392936198475817_xxIEBxuq.jpg

Your request will cost $0.024 per image,
For $1 you can run this model approximately 41 times.

README

HiDream-I1-Full is a 17-billion-parameter open-source text-to-image generation model developed by HiDream.ai. It delivers state-of-the-art image quality across various styles, including photorealistic, cartoon, and artistic renderings. The model achieves top performance on benchmarks such as HPS v2.1, GenEval, and DPG-Bench. Licensed under MIT, it is suitable for both personal and commercial use.

Key Features

  • Exceptional Image Quality: Achieves top scores on HPS v2.1, reflecting superior visual fidelity.
  • Advanced Prompt Alignment: Outperforms other open-source models on GenEval and DPG-Bench benchmarks.
  • Versatile Style Support: Capable of generating images in photorealistic, cartoon, and artistic styles.
  • Open Source & Commercial-Friendly: Released under the MIT license, allowing for unrestricted personal and commercial use.
  • Flexible Deployment Options: Supports full, dev, and fast model variants to accommodate different performance needs.

ComfyUI

hidream-i1-full is also available on ComfyUI, providing local inference capabilities through a node-based workflow, ensuring flexible and efficient image generation on your system.

Limitations

  • Creative Focus: Designed primarily for creative image synthesis; not intended for generating factually accurate content.
  • Inherent Biases: Outputs may reflect biases present in the training data.
  • Input Sensitivity: The quality and consistency of generated images depend significantly on the quality of the input text; subtle variations may lead to output variability.
  • Prompt Dependency: The model's performance is closely tied to the clarity and structure of the prompts; careful crafting may be necessary for optimal results.

Out-of-Scope Use

The model and its derivatives may not be used in any way that violates applicable national, federal, state, local, or international law or regulation, including but not limited to:

  • Exploiting, harming, or attempting to exploit or harm minors, including solicitation, creation, acquisition, or dissemination of child exploitative content.
  • Generating or disseminating verifiably false information with the intent to harm others.
  • Creating or distributing personal identifiable information that could be used to harm an individual.
  • Harassing, abusing, threatening, stalking, or bullying individuals or groups.
  • Producing non-consensual nudity or illegal pornographic content.
  • Making fully automated decisions that adversely affect an individual’s legal rights or create binding obligations.
  • Facilitating large-scale disinformation campaigns.

Accelerated Inference

Our accelerated inference approach leverages advanced optimization technology from WavespeedAI. This innovative fusion technique significantly reduces computational overhead and latency, enabling rapid image generation without compromising quality. The entire system is designed to efficiently handle large-scale inference tasks while ensuring that real-time applications achieve an optimal balance between speed and accuracy. For further details, please refer to the blog post.