Tools0News0Companies0Events0Jobs0Models0Devices0Robots0Tutorials0
AI

GPT Image 2

Multimodal

OpenAI's state-of-the-art image generation and editing model with advanced text rendering and multimodal reasoning.

Paper LinkUse this Model

Model Overview

GPT Image 2 is OpenAI's next-generation multimodal image generation model designed for high-quality image synthesis and editing. It supports text and image inputs, advanced multilingual text rendering, flexible aspect ratios, and reasoning-assisted image generation workflows. The model is available through the OpenAI API and integrated into ChatGPT Images 2.0.

Capabilities

Inputs

Text, image

Outputs

Image

Task categories

Image generation, image editing, creative AI, multimodal reasoning

Languages

Multilingual

Tags

Image-generation, image-editing, multimodal, vision, creative-ai, text-rendering, high-fidelity

Model Details

Developer
OpenAI
Version
gpt-image-2
Release date
April 21, 2026
Model type
Multimodal
Context window
TBA
Open source
No
Commercial use
Allowed
Fine-tunable
No
License
OpenAI proprietary

Architecture

Type
Transformer-based multimodal image generation architecture
Base model
GPT Image 1.5
Active params
TBA
Total params
TBA

Benchmarks

No benchmark scores listed.

Safety & Compliance

Notes
Includes updated safeguards and image provenance protections
Red teamed
Yes
Training
RLHF, supervised fine-tuning
Fine-tunable
No
Base model
GPT Image 1.5
Usage recommendations
TBA