OpenAIOpenAI·🖌️ Inpainting

GPT Image 2

anonymized
Try on Venice.ai ↗
Quick reference
GPT Image 2 — TLDR
  • - 🆕 OpenAI's editing endpoint for the GPT Image 2 family, released April 2026.
  • - 🎯 Supports inpainting and outpainting via an optional mask image.
  • - 🔧 Applies targeted edits while preserving unmasked regions of the picture.
  • - 👁️ Accepts multiple reference images as context for an edit.
  • - 📏 Processes every image input at high fidelity automatically.
  • - 💬 Built for prompt adherence and contextual world knowledge.
  • - 🏢 Exposed as a dedicated image-to-image model id in OpenAI's API.
💰 Pricing
$0.360
per edit
📅 On Venice since
Apr 21, 2026
43 days ago
Provider

OpenAI is an American artificial intelligence research organization headquartered in San Francisco, structured as both a for-profit public benefit corporation and a nonprofit foundation. The lab developed the GPT family of large language models, the DALL-E…

Read full profile →
24 models on Venice
13 text · 4 video · 2 image · 2 embedding · 2 inpaint · 1 asr
Since Jan 15, 2025

About this model

GPT Image 2 is OpenAI's image generation and editing model, and this entry covers its editing (inpainting) endpoint, exposed as an image-to-image interface. You provide one or more reference images alongside a text prompt, optionally with a mask, and the model applies targeted edits while preserving the parts of the picture you did not specify. It supports both inpainting and outpainting and can draw context from multiple input images. OpenAI describes the line as built for developers needing strong prompt adherence and broad world knowledge.

As the successor to GPT Image 1.5, it sits in the same editing family and pairs with the text-to-image sibling GPT Image 2. A notable change versus the prior generation is fidelity handling: for GPT Image 2 the input fidelity parameter is omitted entirely because the model processes every image input at high fidelity automatically, rather than requiring manual tuning.

The editing variant is reached through the same gpt-image-2 model identifier in OpenAI's API, configured for image-to-image work rather than generation from scratch. This makes it suited to localized retouching, extending canvases, and compositing edits guided by reference imagery.

OpenAI also ships text models like GPT-5.5 and video models in the Sora family, but GPT Image 2 Edit is the dedicated editing tool within that lineup.

This About section is AI-generated from public sources (Claude Opus 4.8), with no human editing. It may contain inaccuracies — verify critical details against the sources listed above.

Data sources: Venice API · HuggingFace · Wikipedia — enrichment updated 13h ago