Qwen-Image-2512

Qwen-Image-2512 is the December 2025 update to Qwen-Image, a text-to-image foundational model. It is designed to generate high-quality images from textual prompts with enhanced capabilities in realism, detail rendering, and text integration.

Key Improvements:

  • Enhanced Human Realism: Reduces the "AI-generated" appearance, particularly for human subjects, by improving facial details, skin texture, and environmental context. Accurately captures age-specific features (e.g., wrinkles in elderly portraits) and subtle postures (e.g., leaning forward in a relaxed pose).
  • Finer Natural Details: Renders landscapes, animal fur (e.g., golden retrievers, argali sheep), water flow, and foliage with greater precision and natural texture. Achieves realistic depictions of dynamic scenes (e.g., mist over cliffs, cascading waterfalls). 
  • Improved Text Rendering: Enhances accuracy and layout for textual elements in images (e.g., PPT slides, infographics, and posters). Faithfully integrates text with images, ensuring clarity, proper placement, and stylistic consistency in complex compositions.
  • Evaluated over 10,000 rounds on AI Arena, it outperforms other open-source models and remains competitive with closed-source alternatives.

The model is a component of the image generation pipeline, consisting of:

  • Text encoder: ~8.3B parameters,
  • Transformer: ~20.4B parameters,
  • VAE: ~126M parameters.

Total: ~28.8B parameters


Announce Date: 30.12.2025
Parameters: 20B
Developer: Qwen
Diffusers Version: 0.36.0.dev0
vLLM-Omni Version: 0.12.0rc1
License: Apache 2.0

Public endpoint

Use our pre-built public endpoints for free to test inference and explore Qwen-Image-2512 capabilities. You can obtain an API access token on the token management page after registration and verification.
Model Name Context Type GPU Status Link
There are no public endpoints for this model yet.

Private server

Rent your own physically dedicated instance with hourly or long-term monthly billing.

We recommend deploying private instances in the following scenarios:

  • maximize endpoint performance,
  • enable full context for long sequences,
  • ensure top-tier security for data processing in an isolated, dedicated environment,
  • use custom weights, such as fine-tuned models or LoRA adapters.

Recommended server configurations for hosting Qwen-Image-2512

Prices:
Name GPU Price, hour Generation time, sec.
teslaa100-1.16.64.160 1 $2.37 Launch
h100-1.16.64.160 1 $3.83 Launch
h100nvl-1.16.96.160 1 $4.11 Launch
h200-1.16.128.160 1 $4.74 Launch
Prices:
Name GPU Price, hour Generation time, sec.
teslav100-1.12.64.160 1 $1.20 Launch
rtx5090-1.16.64.160 1 $1.59 Launch
teslaa100-1.16.64.160 1 $2.37 Launch
h100-1.16.64.160 1 $3.83 Launch
h100nvl-1.16.96.160 1 $4.11 Launch
h200-1.16.128.160 1 $4.74 Launch
Prices:
Name GPU Price, hour Generation time, sec.
teslat4-1.16.16.160 1 $0.33 Launch
teslaa2-1.16.32.160 1 $0.38 Launch
teslaa10-1.16.32.160 1 $0.53 Launch
rtx3090-1.16.24.160 1 $0.83 Launch
rtx4090-1.16.32.160 1 $1.02 Launch
teslav100-1.12.64.160 1 $1.20 Launch
rtx5090-1.16.64.160 1 $1.59 Launch
teslaa100-1.16.64.160 1 $2.37 Launch
h100-1.16.64.160 1 $3.83 Launch
h100nvl-1.16.96.160 1 $4.11 Launch
h200-1.16.128.160 1 $4.74 Launch

Related models

Qwen-Image
Qwen-Image-Edit
Qwen-Image-Edit-2511
FireRed-Image-Edit-1.0
FireRed-Image-Edit-1.1

Need help?

Contact our dedicated neural networks support team at nn@immers.cloud or send your request to the sales department at sale@immers.cloud.