Every alternate day there's a new launch.
From Fable 5 to GPT 5.6 to Opus 5 to Grok 4.6 to Fable 5.1 to GPT 6 to Muse to Grok 4.7 to Opus 5.5 to Sonnet 5.5 to GPT 6.1 to Dots. As an AI enthusiast, I am now lost.
In bcherny's video, Sonnet 5.5 and Sonnet 5 fix the same bug in Claude Code
5.5 took 12.3s and 13.3K tokens. 5 took 17.5s and 18.2K tokens Sonnet 5.5 costs $2 input and $10 output per million tokens. Opus 5.5 costs $4 and $20 Which do you leave as
Opus 5.5 dropped at 40% less than Opus 5.
Your safety evals are measuring the wrong thing — turn-1 checks miss turn-12 failures. Big week for the stack.
Since Opus will soon be limited, I tried having Sonnet5.5 create a MV using JavaScript.
Medium because taste doesn't change with effort. I feel like it has more sense than ChatGPT. ChatGPT can be used in chat with Sol thinking performance without a 5-hour limit, so I use it differently with Claude. Easy techno
You have reached the end of the archive
All of Claude Opus 5