- OpenAI has made ChatGPT 4o free for everyone, but should you subscribe to ChatGPT Plus for GPT-4 access?
- In our tests, ChatGPT 4o and ChatGPT 4 performed on par. In fact, benchmarks suggest that the Omni model is a notch better than the GPT-4 model.
- Free users can query up to 10 messages on ChatGPT 4o which resets after every five hours. Paid users have 5x the limit.
OpenAI recently announced its end-to-end multimodal model, ChatGPT 4o, and made it free for everyone. Not just that, free users also get many premium features that were earlier reserved for ChatGPT Plus users. So in this article, we compare ChatGPT 4o and ChatGPT 4 models and find the difference between free ChatGPT and ChatGPT Plus. On that note, let’s begin.
ChatGPT Free vs ChatGPT Plus: Differences
Here are the differences between ChatGPT 4o and ChatGPT 4, available on the free version of ChatGPT and ChatGPT Plus. We have also performed some reasoning tests to understand their capability differences.
ChatGPT Free ChatGPT Plus Models GPT-4o
GPT-3.5GPT-4o
GPT-4
GPT-3.5Internet Access Yes Yes File and Document Upload Yes Yes Analyze data and create charts Yes Yes Upload and analyze images Yes Yes Access Custom GPT Yes Yes Create Custom GPT No Yes Explore GPT Store Yes Yes Memory Yes Yes Message limits 10 queries on GPT-4o (subject to system load)
Resets every 5 hours
Rest of the query on GPT-3.55x more message limit than free users on GPT-4o Voice Conversation Old Voice Mode Natural Voice Conversation (Upcoming) Access to upcoming frontier models No Yes
1. Find Drying Time
In our first test, ChatGPT 4o and ChatGPT 4 performed along the same lines. Despite having access to Code Interpreter, none of the models used it for mathematical calculation and straight up answered using logical reasoning.
If it takes 1 hour to dry 15 towels under the Sun, how long will it take to dry 20 towels?
Winner: ChatGPT 4o and ChatGPT 4
![reasoning test between chatgpt 4o and chatgpt 4](https://beebom.com/wp-content/uploads/2024/05/reasoning-test-between-chatgpt-4o-and-chatgpt-4.jpg?w=995&quality=75)
2. The Elevator Test
In the second reasoning test, ChatGPT 4o and ChatGPT 4 answered correctly and came to the correct conclusion. Both models answered floor 4.
There is a tall building with a magic elevator in it. When stopping on an even floor, this elevator connects to floor 1 instead.
Starting on floor 1, I take the magic elevator 3 floors up. Exiting the elevator, I then use the stairs to go 3 floors up again.
Which floor do I end up on?
Winner: ChatGPT 4o and ChatGPT 4
3. Find the Weight
In the next test which has proved to be a challenge for a lot of LLMs, both ChatGPT 4o and ChatGPT 4 aced it without any issues. Both models say “A kilo of feathers is heavier than a pound of steel.” In our recent comparison between ChatGPT 4o and Gemini 1.5 Pro, Google’s AI model failed to correctly answer this question.
What's heavier, a kilo of feathers or a pound of steel?
Winner: ChatGPT 4o and ChatGPT 4
4. Follow User Instructions
I asked ChatGPT 4o and ChatGPT 4 to generate 10 sentences that end with the word “deep learning” and both models got it right 10 out of 10. In properly following instructions, ChatGPT 4o and GPT-4 join Llama 3 70B in understanding the user intent and showcasing great alignment.
Generate 10 sentences that end with the word "deep learning"
Winner: ChatGPT 4o and ChatGPT 4
5. The Apple Test
I ran the final question to decide whether both models exhibit similar levels of intelligence. And indeed, ChatGPT 4o and ChatGPT 4 both gave correct answers with clear reasoning. Kudos to OpenAI for making the Omni model 2x faster than GPT 4, yet offering the same level of intelligence.
I have 3 apples today, yesterday I ate an apple. How many apples do I have now?
Winner: ChatGPT 4o and ChatGPT 4
Closing Thoughts
After playing with both models, I am assured that ChatGPT 4o is indeed a GPT-4 class model. Both perform intelligently and are quite similar in their reasoning and alignment. In fact, OpenAI’s benchmark results suggest that ChatGPT 4o is a notch better than the ChatGPT 4 model. LMSYS Leaderboard also indicates the same.
ChatGPT 4o scored 88.7 on the MMLU benchmark, and the latest GPT-4 (gpt-4-turbo-2024-04-09) scored 86.5. The trend is similar across HumanEval, MATH, and GPQA benchmark tests. What I find more potential-worthy than their capabilities is the speed at which ChatGPT 4o operates. It’s 2x faster, and 50% cheaper than GPT-4 which is remarkable.
As for free ChatGPT users, I would say, a limit of 10 messages every five hours is pretty good. You can access the state-of-the-art ChatGPT 4o model at no cost. In addition, you get a lot of premium features. Learn how to use ChatGPT 4o right now for free.
That said, if you are a power user and frequently use ChatGPT for your everyday work, getting a subscription would be desirable. I recently got ChatGPT 4o access on my free account, but its performance was not up to the mark. So for advanced users, go ahead and get the ChatGPT Plus subscription.
How do you use chat-gpt 4turbo on the free version.. isn’t that still 3.5?
I need new articles on ChatGPT like this please consider my request, please suggest courses which are best for using ChatGPT more efficiently it will help a lot…