GPT Image 1.5

Note: As of July 2026, GPT Image 1.5 is considered a deprecated model by OpenAI, having been superseded by newer models like GPT Image 2.

Model Overview

GPT Image 1.5 (gpt-image-1-5) was released by OpenAI in December 2025 as an iterative upgrade to the GPT Image 1 family. It continued to leverage a natively multimodal, autoregressive architecture but brought substantial improvements in editing precision, text rendering accuracy, and generation speed. The model was specifically optimized for developers requiring granular control over image modifications and high-fidelity text integration within their applications.

Capabilities

• Precision Editing: Allowed for highly accurate region-specific inpainting and modifications. It could blend elements seamlessly while preserving original lighting, composition, and facial consistency.
• Enhanced Text Rendering: Delivered significantly higher accuracy in rendering readable text, logos, and labels compared to the baseline GPT Image 1.
• High-Speed Generation: Operated up to 4x faster than its predecessor, greatly reducing latency for end-users and real-time applications.
• Extended Context: Featured a 4,000-token context window to accommodate highly detailed and complex multi-part prompts.

Example Use Cases

• Typography and Branding: Generating logos, signage, and marketing materials that require accurate and readable text.
• Photographic Retouching: Precisely modifying specific regions of an image (e.g., changing clothing or adjusting background elements) without altering the subject's identity or the scene's global lighting.
• High-Throughput Applications: Applications requiring rapid image generation, benefiting from its substantial speed increase.

Performance & Benchmarks

At its peak, GPT Image 1.5 was recognized for its exceptional speed improvements (up to 4x faster than GPT Image 1) and its robust 4,000-token context window. Its text-rendering accuracy set a new standard for AI image generators before the release of the next-generation models in 2026.

Intended Use & Limitations

The model was designed for API-driven workflows demanding high speed and precise editing capabilities. However, because it is now deprecated, developers are highly encouraged to migrate to GPT Image 2 to take advantage of current state-of-the-art performance and ongoing support.

About OpenAI

OpenAI is a leading AI research and deployment company known for pioneering large language models and multimodal AI systems, including the GPT and DALL-E series, with the mission to ensure AGI benefits all of humanity.