Everyone thought it would just be faster or smarter.
But what OpenAI just dropped might actually change how you work forever.
Technical Benchmarks
As usual, let’s check technical benchmarks.
GPT-5 sets the bar high in software engineering and multilingual code editing. It beats all GPT models across SWE-bench and Aider Polyglot tasks.
In visual problem-solving and math competitions, GPT-5 is good.
100% accuracy in AIME 2025 with thinking mode is really interesting and will make us question the educational system in the future.
GPT-5 is significantly more reliable than others.
It hallucinates less, answers more accurately, and reduces error rates in traffic-like prompts by over 75%.
Reliability and accuracy benchmark
As you can see, OpenAI did not compare with other models as they did, but just compared with their previous models.
Use-Cases
Let’s see GPT-5 in action.
What is the Bernoulli effect?
In this use case, GPT-5 breaks down the complex concept of the Bernoulli principle with clarity.
What are Bernoulli models? Answered by GPT-5
Visual Implementation of Bernoulli Models in Canvas
Next, the ChatGPT team wants to create an SVG that explains Bernoulli models in canvas.
Visual Implementation of Bernoulli Models
Writing in Natural Tone
At this step, the OpenAI team tested the writing feature of GPT-5 and observed a more natural tone.
Writing with GPT-5 in natural tone
Building a Web App
Now, GPT-5 builds a web app like Claude, and you can test the Web APP in Canvas, similar to Claude's artifact.
In this use case, the OpenAI team created a web app to learn French.
ChatGPT in Health -Doing Diagnosis
This real use case melted hearts. A woman used GPT to understand her cancer report. I hope she’s recovering well!
New Models in API
There are three models for different purposes with different prices.
GPT-5 nano is the cheapest while GPT-5 looks more powerful.
More flexibility for developers and product teams.
Coding Ability
GPT-5 performs better in coding tests, including SWE-bench and multilingual tasks.
It’s now more accurate and context-aware.
Long-Context Performance
Massive context window support.
It keeps track of longer threads better than ever, ideal for large documents and logs.
GPT-5 On Cursor
At this point, they demonstrate GPT-5's capabilities in vibe-coding with Cursor. In this presentation, they focused on demonstrating their coding capabilities on Cursor. They also tested the GPT with the CEO of Cursor, who stated that this is the best model they have tested.
Finance Dashboard For Start-up
Developing Finance Dashboard with Cursor via GPT-5
Here, they create a finance dashboard using the GPT-5 API via a cursor. Here is the result.
Here is the prompt, by the way.
Create a Game with Cursor
At this step, the OpenAI team created a game with Cursor.
Here is the result.
You can chat with the characters who are walking in the castle.
Castle Game’s first presentation done by GPT-5
And they were playing a game during the presentation!
Final Thoughts
I’ll test ChatGPT-5 more in-depth soon, but this is the first analysis now. Follow me to see similar kinds of articles in the upcoming days.
If you're looking for more, we offer Substack, a dedicated team that creates content for learning AI, and a platform featuring no-code AI tools.
Here are free resources for you to take advantage of.
Here is the ChatGPT cheat sheet.
Here is the Prompt Techniques cheat sheet.
Here is my NumPy cheat sheet.
Here is the source code of the “How to be a Billionaire” data project.
Here is the source code of the “Classification Task with 6 Different Algorithms using Python” data project.
Here is the source code of the “Decision Tree in Energy Efficiency Analysis” data project.
Here is the source code of the “DataDrivenInvestor 2022 Articles Analysis” data project.
“Machine learning is the last invention that humanity will ever need to make.” Nick Bostrom
Originally published at https://www.learnwithmeai.com/p/gpt-5-is-finally-released-here-is.