GPT 5.2 has just been released. Let's discover together!
The most powerful model in the world (as they said.)
Yesterday, OpenAI dropped GPT 5.2. The crown changed hands again.
This keeps happening now. The throne never stays warm.
So I tested it. Three rounds: writing, coding, reasoning. GPT 5.2 versus Claude.
The judge? Gemini 3.0 Pro. Let the games begin.
Writing
Claude writes better. At least, that’s my bet.
Time to prove it. Here’s the prompt:
Writing Prompt
My article title is: ‘GPT-5.2 has just been released. Let’s discover it together!’ Write an introduction. Start with a hook. Next, explain what the article will include. Finish with curiosity.
Claude & ChatGPT
Claude searched the web first, but ChatGPT writes without browsing.
Both delivered. But Claude went deeper. More detail.
Let’s compare them using Gemini 3.0 Pro, using this prompt.
Evaluation Prompt
You are a writing evaluator. Below are two texts written by two different AI models on the same topic.
Text Ch:
[PASTE TEXT A HERE]Text Cl:
[PASTE TEXT B HERE]Compare these two texts only in terms of writing quality. Evaluate clarity, tone, structure, flow, and readability for a general audience. Explain the strengths and weaknesses of each text. Then conclude which text is stronger overall and why. Focus purely on writing, not on factual accuracy or technical depth.
I did not give their exact names (Claude, ChatGPT) to avoid bias. Let’s see Gemini’s evaluation.
Claude 1 - ChatGPT 0.
LearnAIWithMe Paid Perks
The standard subscription ($12/mo or $100/year) gives you full access to all my articles and posts on Substack
The Founding Member tier ($200/year) includes everything above PLUS access to The Vault, 700+ prompts, projects, AI tools, and AI labs.