Head-to-Head

Claude 3.5 Sonnet vs ChatGPT-4o: The Ultimate Showdown

Which LLM reigns supreme in 2024? We test coding, creative writing, and reasoning.

Quick Summary

Advantages

  • Claude: Better coding & writing
  • ChatGPT: Better multimodal & tools

Disadvantages

  • Claude: No web browsing (native)
  • ChatGPT: More 'lazy' responses recently

The Contenders

Claude 3.5 Sonnet by Anthropic vs GPT-4o by OpenAI.

Coding Test

In our Python refactoring tests, Claude 3.5 Sonnet consistently produced cleaner, more modern code and explained its logic better. GPT-4o was faster but prone to hallucinating libraries occasionally.

Creative Writing

Claude feels more human. ChatGPT still has that "AI accent" (overusing words like 'tapestry' and 'testament').

Multimodal

GPT-4o's voice mode and vision capabilities are slightly ahead in terms of speed and integration, especially with the native app.

Winner?

For coding and writing: Claude. For general assistance and voice: ChatGPT.