73 Claude Opus 4.5 vs GPT 5.2 Extra High
These two companies' strongest models, which one is actually stronger?
Yesterday I pushed claude code hard all day, conclusion reached at night:
Long story short, look at the pictures directly
This is claude code at full power, model all configured with claude opus 4.5, thinking budget maxed out, prompt with ultrathink result
This is codex, gpt 5.2 extra high result
Development task is troubleshooting a problem and correctly parsing SSE response, not particularly complex, task difficulty medium-high
Purchase required to continue
This is a paid article. After signing in, your purchase will be unlocked automatically.
No comments yet. Be the first to share your thoughts.