alt Hacker NewsThis looks cherry-picked, for example Claude Opus had a higher score on SWE-Bench Verified so they conveniently left it out, also GDPval is literally a benchmark made by OpenAI
Replies
minadotcom • last Thursday at 9:24 PM
agreed.
This looks cherry-picked, for example Claude Opus had a higher score on SWE-Bench Verified so they conveniently left it out, also GDPval is literally a benchmark made by OpenAI
agreed.
And who believes that the difference between 91.9% and 92.4% is significant in these benchmarks? Clearly these have margins of error that are swept under the rug.