A simple test, but its result is interesting 🐿️
Claude Opus 5.5 and GPT-6 Astra were asked to perform the same task: a squirrel launches a chestnut using a catapult into the sky. Same description, one attempt per model. Result in terms of cost and time: Claude Opus 5.5: about $3.30 and 23 minutes. GPT-6 Astra:
I've been editing videos for 7 years.
Now that I use SpaceCutAI, Opus is a much better editor than me A full animation in 5 minutes vs hours for a human, you can't compete with AI from Billionerds podcast with eeelistar
These 19 GitHub repos let Opus 5.5 create 3D animations, motion graphics, voiceovers, and complete videos without After Effects.
Send them to Opus 5.5 and turn your agent into a motion design studio. here are the 19 repositories: motion design & video creation 1.
This is all you need to know about Science's Last Exam (tl;dr)
113 executable research tasks across 9 fields (34 institutions). Best overall score is 37.5/100 Opus 5. 36.15/100 GPT-6.1 Sol. All 8 frontier models scored 0 on true end-to-end research.
You have reached the end of the archive
All of Claude Opus 5