Anthropic re-wrote the Sonnet 5 story post-launch. The first BrowseComp cost-performance chart showed Sonnet 5 lagging Opus 4.8. The replacement chart paints... Read More...
Source
source & further reading
vincentschmalbach.com — original article
Anthropic Merges 30,000 Changes a Day, and It Shows
Tests Come for Free Now, and Most of Them Are Pointless
OpenAI Does Not Trust Its Own Models, and Rightly So