Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · Efficient-Large-Model

SANA1.5_1.6B_1024px_diffusers

SANA1.5_1.6B_1024px_diffusers is a 1.6B-parameter text-to-image model from Efficient-Large-Model. According to the model card, it uses BF16 precision to generate 1024px-based images through a Diffusers SanaPipeline.

Publisher
Efficient-Large-Model
Task
text-to-image
Model type
Unknown
License
apache-2.0
Library
sana
Publication status
Accepted · not indexed

Model overview

SANA1.5_1.6B_1024px_diffusers is published by Efficient-Large-Model as a text-to-image release. Hub metadata records the sana library with diffusers, safetensors, SANA-1.5, and 1024px tags. According to the model card, it is a scalable Linear-Diffusion-Transformer generative model with 1.6B BF16 parameters for multi-scale 1024px image generation.

Recorded capabilities

1.6B BF16 diffusion model

According to the model card, the model has 1.6B parameters in torch.bfloat16 precision.

1024px multi-scale output

The publisher describes this release as developed for 1024px-based images with multi-scale height and width.

Training and inference scaling account

According to the model card, SANA-1.5 covers 1.6B-to-4.8B growth, depth pruning, VLM-selection inference scaling, and GenEval and DPGBench results without captured numeric tables in this evidence.

Diffusers example

The card provides a SanaPipeline snippet generating a 1024x1024 image at guidance scale 4.5 over 20 inference steps.

Use cases in the source record

  • 1024px text-to-image generation with the documented Diffusers SanaPipeline, guidance, and inference-step settings.
  • Research use of the linked Sana repository with advanced diffusion samplers such as Flow-DPM-Solver.

Limitations and unknowns

  • Performance claims for GenEval and DPGBench are publisher statements without captured numeric tables in this evidence.
  • According to the model card, factual or true depictions of people or events are out of scope because the model was not trained for factual representation.
  • The Diffusers integration is marked as under construction in the captured card text.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Efficient-Large-Model/SANA1.5_1.6B_1024px_diffusers

Captured: Unknown. Processed: 2026-09-07T19:35:05.043218+00:00.

🐱 Sana Model Card Model We introduce SANA-1.5 ，an efficient model with scaling of training-time and inference time techniques. SANA-1.5 delivers: efficient model growth from 1.6B Sana-1.0 model to 4.8B, achieving similar or better performance than training from scratch and saving 60% training cost; efficient model depth pruning , slimming any model size as you want; powerful VLM selection based inference scaling , smaller model+inference scaling > larger model; Top-notch GenEval & DPGBench results. Detailed results are shown in the below table. Source code is available at https://github.com/NVlabs/Sana . Model Description Developed…

F001F002F003F004F005F006F007F008F009F010F011F012F013F015F016F017