Short answer
ChatGPT image generation may take a few minutes, depending on how complex your request is. OpenAI’s help article does not give an exact time in seconds or promise when an image will be ready, so treat “a few minutes” as guidance—not a guarantee. (OpenAI Help Center)
You can keep using ChatGPT while the image is being created. OpenAI also says its ChatGPT Images model generates images faster than its predecessor, but does not specify a duration or the size of that speed improvement. (OpenAI Help Center)
Why do reported times vary?
A fixed estimate can be misleading: requests differ in complexity, and service conditions can change. Older third-party timing reports should not be treated as reliable forecasts for today’s ChatGPT experience. For example, AI Free API reported an overall range of 10–60 seconds in its July 11, 2025 article, based on GPT-4o-era image generation. Those are historical third-party estimates, not current ChatGPT performance figures. (AI Free API)
If an image is taking longer than you expected, give it some time. If it appears stuck or fails, check OpenAI’s status page for a service issue, then try again if the service is operating normally.
ChatGPT or an API for production?
For occasional image creation, the ChatGPT interface is the straightforward choice: enter your prompt and wait for the result. For an app, automated workflow, or batch process, an image-generation API may be worth evaluating—but using an API does not by itself prove that each image will generate faster. Parallel requests may help a system handle more work at once; that is about throughput, not necessarily the time to complete one image.
Before choosing an API for production:
- Set a target: Decide how long a single image may take and how many requests your workflow needs to handle.
- Check the current API documentation: Confirm the available image-generation options and applicable limits for your intended use. Start with OpenAI’s image-generation guide.
- Measure your workload: Test representative prompts and settings in the intended environment. Record elapsed time from submission until the image is ready, and compare typical and slower results with your target.
- Choose based on the results: Keep the interactive ChatGPT workflow if it meets your needs; use an API workflow if its automation and measured performance fit your requirements.
Bottom line: OpenAI says image generation may take a few minutes, depending on request complexity. That is not a maximum or a guaranteed completion time. The cited help article gives no exact time in seconds. For production, benchmark the actual API workflow rather than assuming it will be faster.
Sources
Images in ChatGPT — OpenAI Help Center