OpenAI has rolled out GPT-4.1 and GPT-4.1 Mini directly into ChatGPT. But are they worth the hype? I put them to the test with real-world promptsāand the results were surprising.
- Instruction Following: GPT-4.1 excels at following detailed prompts, making it ideal for agentic workflows. However, it requires well-structured prompts to shine.
- Performance: In side-by-side comparisons, GPT-4.1 was faster and more efficient than GPT-4 Mini High, delivering markdown-formatted reports swiftly.
- Coding Capabilities: While GPT-4.1 handled data analysis impressively, it struggled with complex coding tasks, sometimes failing to complete them.
- Prompting Guide: OpenAI's extensive prompting guide emphasizes the need for explicit instructions, planning, and reflection to get the best results from GPT-4.1.
What do you think?