ChatGPT launched on November 30, 2022, as a free research preview by OpenAI

OpenAI released ChatGPT to the public on November 30, 2022. The tool was available free through a web interface at openai.com, and it reached one million users in five days — faster than any consumer application at that time. The initial version was built on OpenAI's GPT-3.5 model, a large language model trained on text data up to early 2022.

The launch was not a surprise announcement. OpenAI had been developing large language models for years, and researchers and developers already knew the company was working on conversational AI. What made the November 2022 release significant was that it put a working version directly in the hands of the general public, not just researchers or paying customers.

Key Takeaways

  • ChatGPT became publicly available on November 30, 2022, built on the GPT-3.5 model and offered free through OpenAI's website.
  • GPT-4, a more capable model, launched on March 14, 2023, initially available only to ChatGPT Plus subscribers for a monthly fee.
  • Mobile apps for iOS and Android arrived in May 2023 and December 2023, extending access beyond web browsers.
  • Voice and image features rolled out gradually through 2023 and 2024, allowing users to speak to ChatGPT and upload pictures for analysis.
  • GPT-4 Turbo, a faster and cheaper version of GPT-4, launched in November 2023 for subscribers.

The first major upgrade: GPT-4 in March 2023

OpenAI released GPT-4 on March 14, 2023, about three and a half months after the initial launch. GPT-4 was a substantially more capable model than GPT-3.5. It could handle longer text inputs, reason through more complex problems, and produce fewer factual errors. However, it was also slower and more expensive to run.

GPT-4 was not immediately free. OpenAI made it available only to ChatGPT Plus subscribers, who paid $20 per month. The free version of ChatGPT continued to use GPT-3.5. This two-tier system remains in place: free users access GPT-3.5, while paid subscribers can choose between GPT-3.5 and GPT-4.

Mobile apps and expanded access through 2023

The web interface was ChatGPT's only way to access the tool for the first six months. OpenAI released an iOS app on May 18, 2023, allowing iPhone users to access ChatGPT without a browser. The Android app followed much later, on December 13, 2023. Both apps included the same features as the web version but were optimized for phone screens and touch input.

During this same period, OpenAI began rolling out additional features. Voice input arrived in September 2023, letting users speak to ChatGPT instead of typing. Image upload capability launched in September 2023 as well, allowing users to share pictures and ask ChatGPT questions about them. These features were not available to all users at once; OpenAI rolled them out gradually to different groups.

GPT-4 Turbo and faster responses in late 2023

On November 6, 2023, OpenAI announced GPT-4 Turbo, a version of GPT-4 designed to be faster and cheaper than the original. GPT-4 Turbo could also handle longer text inputs — up to 128,000 tokens, roughly equivalent to 100,000 words. This made it useful for tasks like analyzing entire documents or long conversations in a single request.

GPT-4 Turbo was available only to ChatGPT Plus subscribers. The pricing per token (the unit OpenAI uses to measure input and output) was lower than GPT-4, making it a more economical choice for users who needed speed or wanted to process large amounts of text. Free users continued to access GPT-3.5.

Vision and multimodal features in 2024

Throughout 2024, OpenAI continued adding capabilities to ChatGPT. GPT-4 Vision, which could analyze images in detail, became available to all ChatGPT Plus subscribers. Users could upload screenshots, photographs, diagrams, and charts, and ChatGPT could describe them, read text within them, or answer questions about their content.

OpenAI also released GPT-4o (the "o" stands for "omni") in May 2024, a model designed to handle text, images, and audio more naturally. GPT-4o was faster than previous versions and became available to free users on a limited basis, with more generous usage limits for paid subscribers. This represented a shift toward making advanced capabilities available to the free tier, though with restrictions on how many requests users could make per day.

What version are you using now

If you use ChatGPT today, which model you access depends on your account type. Free users access either GPT-3.5 or GPT-4o, depending on OpenAI's current allocation and server load. ChatGPT Plus subscribers can choose between GPT-3.5, GPT-4, GPT-4 Turbo, and GPT-4o from a dropdown menu at the top of the chat window.

You can see which model you are using by looking at the top of the chat interface. If you are on the web version, the model name appears next to the chat title. On mobile apps, it may appear in the settings or at the top of the conversation. If you are unsure whether you have a free or paid account, check your account settings — paid accounts show an active subscription with a renewal date.

Why the version history matters for your work

Understanding when features launched helps explain why ChatGPT might not do something you expected. If you read an article from 2023 describing ChatGPT's capabilities, some features mentioned may not have existed when that article was written. Image analysis, for example, was not available until September 2023, so any guide written before that date would not mention it.

The version you are using also affects what ChatGPT can do. GPT-3.5 is faster and uses less computing power, but it makes more mistakes on complex reasoning tasks. GPT-4 and GPT-4o are more accurate but slower. If you are a free user and a task is not working well, switching to a paid account and selecting GPT-4o might solve the problem — or it might not, depending on what you are asking the tool to do.

Frequently Asked Questions

Is ChatGPT still free to use?

Yes. OpenAI offers a free tier that includes access to ChatGPT with GPT-3.5 and limited access to GPT-4o. Free users can make a certain number of requests per day before hitting a usage limit. ChatGPT Plus, the paid subscription at $20 per month, removes these limits and provides access to all available models.

What is the difference between GPT-3.5 and GPT-4?

GPT-4 is more accurate at complex reasoning, math, and coding tasks. It makes fewer factual errors and can handle longer text inputs. GPT-3.5 is faster and uses less computing power. For everyday writing and questions, both work well; the difference becomes noticeable on specialized or difficult tasks.

Can I use ChatGPT on my phone?

Yes. OpenAI offers free iOS and Android apps that work with both free and paid accounts. The mobile apps have the same features as the web version, including voice input and image upload. You can download them from the Apple App Store or Google Play Store.

When did ChatGPT get the ability to see images?

Image upload and analysis became available in September 2023. This feature was rolled out gradually, so some users received it before others. Today, all ChatGPT users can upload images, though the quality of analysis may vary depending on which model you are using.

Does ChatGPT know about events after 2022?

The original GPT-3.5 model was trained on data through early 2022, so it had no knowledge of events after that date. GPT-4 and GPT-4o have training data through April 2024, so they know about more recent events. However, neither model receives real-time information, so they cannot tell you what is happening today.