FireRed-Image-Edit-1.0

A model for image editing tasks that ensures high accuracy, quality, and consistency across various scenarios.
Key features:

• High performance: Achieves leading results among open-source solutions in precise instruction following, image quality, and visual consistency.
• Text style preservation: High-quality retention of text styles in images.
• Old photo restoration: High-quality enhancement and restoration of old photographs.
• Multi-reference editing: Supports editing of multiple images (e.g., "virtual try-on").

---

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

• Text encoder: ~8B parameters,
• Transformer: ~20B parameters,
• VAE: ~126 parameters,

Total: ~28B parameters

Announce Date: 14.02.2026
Parameters: 20B
Developer: FireRedTeam
Diffusers Version: 0.36.0.dev0
vLLM-Omni Version: 0.12.0rc1
License: Apache 2.0

Use our pre-built public endpoints for free to test inference and explore FireRed-Image-Edit-1.0 capabilities. You can obtain an API access token on the token management page after registration and verification.

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.

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