NexaAI/sdxl-base
By NexaAI · ↓ 11 · ♥ 0 · Updated 2025-12-15
What's in this repo
This page lists the files in NexaAI/sdxl-base, including model.onnx and more.
- 5 ONNX models (.onnx files)
- 27 files, including 16 .json, 5 .onnx, 2 .txt, 1 .md, 1 .manifest, 1 .data
- Last updated 2025-12-15
ONNX files
| File | Size | Download |
|---|---|---|
| text_encoder/model.onnx | — | Download |
| text_encoder_2/model.onnx | — | Download |
| unet/model.onnx | — | Download |
| vae_decoder/model.onnx | — | Download |
| vae_encoder/model.onnx | — | Download |
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
Stable-Diffusion-XL-Base-1.0
How to run
Visit sdk.nexa.ai/model
Model Description
Stable Diffusion XL Base 1.0 (SDXL 1.0) is a foundation text-to-image model released by Stability AI.
It is the flagship successor to Stable Diffusion 2.1, designed for photorealism, artistic flexibility, and high-resolution generation.
SDXL 1.0 is a latent diffusion model trained on a broad dataset of images and captions. Compared to prior versions, it improves prompt alignment, visual coherence, and output quality, especially in complex scenes and detailed compositions.
Features
- High fidelity image generation: sharper details and improved realism.
- Flexible style range: from photorealistic renders to artistic illustration.
- Better prompt alignment: improved understanding of nuanced or multi-concept prompts.
- High resolution support: natively trained for 1024×1024 images.
- Compositional strength: more accurate handling of multiple subjects and fine object placement.
Use Cases
- Creative content generation (illustrations, art, concept design)
- Product mockups and marketing visuals
- Character and environment ideation
- Storyboarding and visual storytelling
- Research in generative imaging
Inputs and Outputs
Input:
- Text prompts (descriptions, concepts, artistic directions)
- Optional negative prompts to avoid undesired elements
Output:
- Generated image(s) matching the prompt
- Default resolution: 1024×1024 pixels
How to use
1) Install Nexa-SDK
Download and follow the steps under "Deploy Section" Nexa's model page: Download Windows SDK
2) Get an access token
Create a token in the Model Hub, then log in:
nexa config set license '<access_token>'
3) Run the model
Running:
nexa infer NexaAI/sdxl-base
License
- Licensed under: CreativeML Open RAIL++-M License
References
Files (27)
Related models
More from NexaAI
More models
No task type declared.