# High-resolution image generation

FLUX.1 [dev] is a rectified flow transformer with 12 billion parameters, capable of generating high-quality images based on text descriptions, providing powerful image generation capabilities for developers and creative professionals.

simpletuner-lora is a text-to-image and image-to-image conversion model based on PEFT LoRA, derived from the FLUX.1-dev model.

LyCORIS adapter based on Stable Diffusion 3.5 Medium, specializing in photorealistic image generation

A LyCORIS adapter based on FLUX.1-dev, focusing on text-to-image and image-to-image tasks, supporting multiple resolution outputs.

This is a standard PEFT LoRA derivative model based on FLUX.1-dev, focusing on text-to-image and image-to-image generation tasks.

LyCORIS adapter based on HiDream-I1-Full, focusing on high-quality image generation

AuraFlow v0.3 is a fully open-source flow-based text-to-image generation model that supports multiple aspect ratios, with resolutions up to 1536 pixels.

Standard PEFT LoRA model based on FLUX.1-dev, specializing in generating high-quality female character images

A PEFT LoRA model trained based on FLUX.1-dev, focused on text-to-image generation tasks, supporting specific artistic style conversion.

DC-AE is a novel autoencoder architecture designed to accelerate high-resolution diffusion models. It maintains reconstruction quality at high spatial compression ratios through residual autoencoding and decoupled high-resolution adaptation techniques.

DC-AE is a novel autoencoder architecture designed to accelerate high-resolution diffusion models, addressing reconstruction accuracy issues under high compression ratios

Sana is an efficient text-to-image framework capable of generating images with resolutions up to 4096×4096, featuring rapid synthesis of high-resolution, high-quality images.

Sana is an efficient text-to-image framework capable of generating images with resolutions up to 4096×4096, supporting multilingual input.

A text-to-image diffusion model fine-tuned based on FLUX.1-dev, specializing in GArt style image generation

Sana is an efficient text-to-image framework capable of generating images up to 4096×4096 resolution, deployable on laptop GPUs.

A text-to-image model based on LoRA technology, specifically designed for generating e-book cover designs

MLX framework version optimized from Stable Diffusion 3.5 Large, specifically designed for Apple chip-optimized text-to-image generation models

A LoRA fine-tuned model based on Stable Diffusion 3.5, specialized in generating images with a futuristic bronze color style.

Meissonic is a non-autoregressive masked image modeling text-to-image model capable of generating high-resolution images, specifically designed to run on consumer-grade GPUs.

CogView3-Plus-3B is the DiT version of CogView3, supporting text-to-image generation from 512 to 2048 pixels.

An early release version based on Stable Diffusion XL, focusing on generating anime-style illustrations through text-to-image modeling

MLX implementation of Stable Diffusion 3 Medium, focused on text-to-image generation

Hed ControlNet checkpoint specifically designed for the FLUX.1-dev model, for image generation tasks

FLUX.1-dev ControlNet is a deep ControlNet checkpoint developed by Black Forest Labs, suitable for image generation tasks at 1024x1024 resolution.

Canny edge detection control network checkpoint for the FLUX.1-dev model, suitable for 1024x1024 resolution image generation

AuraFlow v0.3 is a fully open-source flow-based text-to-image generation model that supports multiple aspect ratios, with resolutions up to 1536 pixels.

IP adapter for the FLUX.1-dev model, supporting image processing similar to text for text-to-image generation tasks

The FLUX.1-dev ControlNet Collection provides three pre-trained models (Canny edge detection, HED edge-aware, Midas depth map), optimized for 1024x1024 resolution image generation.

Stable Diffusion XL (SDXL) 1.0 is a powerful text-to-image model capable of generating high-quality images from text descriptions.

Lumina-Next-SFT is a 2-billion-parameter Next-DiT model that uses Gemma-2B as the text encoder and is enhanced through high-quality supervised fine-tuning (SFT) for text-to-image generation.

A fully fine-tuned image generation model based on ptx0/pixart-900m-1024-ft-large, specializing in high-quality image generation

Colorful XL is a text-to-image generation model based on stable diffusion technology, capable of producing high-quality and diverse images from text descriptions.

A text-to-image generation model based on Amber XL Epsilon rev1, optimized for selected artist works and specific series/game-related images

A powerful control network model capable of generating high-resolution images with visual quality comparable to Midjourney, achieving precise control through Canny edge detection.

A full-rank fine-tuned text-to-image generation model based on terminus-xl-velocity-v1, supporting multiple resolution outputs

PixArt-Σ is a latent diffusion model based on the Transformer architecture, capable of generating high-resolution images (up to 4K) directly from text prompts.

A LoRA model fine-tuned based on Stable Diffusion XL 1.0, specializing in generating arcane magic-style fantasy concept art images

An image generation model fine-tuned based on the Stable Diffusion 1.5 framework, optimized for high-resolution output stability and supports multiple aspect ratio image generation

Deliberate 2 is a text-to-image generation model that supports generating images in styles such as general, anime, and art.

A text-to-image generation model based on stable diffusion technology, capable of producing high-quality images from text descriptions.