← Models

GPT-Image-1

OpenAI · GPT Image

OpenAI image generation model for prompt-driven creation and iterative editing workflows.

Part of GPT Image family · Other versions: GPT Image 2 , gpt-image-1-mini , GPT Image 1.5
Type
image
Context
N/A
Max Output
N/A
Status
current
API Access
Yes
License
proprietary
image-generation editing multimodal creative api
Released April 2025 · Updated April 4, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on April 4, 2026.

GPT-Image-1 is OpenAI’s image generation model for API and product workflows that need image creation and editing from prompts. It is aimed at teams building visual features rather than purely text assistants.

Capabilities

The model supports text-to-image style generation and iterative image refinement patterns. It is useful for concept visuals, campaign assets, and embedded product generation tools.

Technical Details

Token-centric context/output fields are not directly meaningful for this image model. In this repository, contextWindow and maxOutput are set to 0 intentionally and should be interpreted as N/A for token displays.

Pricing & Access

GPT-Image-1 is exposed through OpenAI API model surfaces. OpenAI’s current pricing page lists text-prompt input at 5.00per1Mtokens,imageinputat5.00 per 1M tokens, image input at 10.00 per 1M tokens, and image output at 40.00per1Mtokens.OpenAIalsocontinuestoframeper−imagecostsinpracticalterms,withsquare−imageoutputcostingroughly40.00 per 1M tokens. OpenAI also continues to frame per-image costs in practical terms, with square-image output costing roughly 0.01, 0.04,or0.04, or 0.17 depending on quality.

Best Use Cases

Use GPT-Image-1 for image ideation services, creative tooling prototypes, marketing asset workflows, and product experiences requiring fast visual generation loops.

Comparisons

Compared with Imagen 4, GPT-Image-1 differences are often ecosystem and quality-style dependent. Compared with Midjourney-style tools, GPT-Image-1 is more API-oriented for product integration. Compared with GPT-image-1.5, the current base model is cheaper to adopt but no longer OpenAI’s newest image route.