Is the new Sol a reply to Sonnet 5.5?

Two mid-tier models, one day apart. Coincidence, or a rivalry playing out in public?

The lineups actually line up. Anthropic's Sonnet is the middle tier, quicker and cheaper than the flagships, just like OpenAI's Sol. Opus and Fable play the flagship role, the counterpart to Astra. And OpenAI's Luna sits at the small, cheap end.

Sonnet 5.5 came out on 28 September. GPT-6.1 Sol followed on the 29th. It looks like a reaction. It almost certainly is not. Nobody goes from idea to release in a day, so both were in the works long before either launched.

Ok, so if it is not a reaction, why does it feel like one?

Because both companies are running the same play. Push the middle tier close enough to the flagship that most people never need to pay flagship prices.

We have seen this movie before. Sonnet 5 came close to the previous Opus on agentic tests for a fraction of the cost, and before that Sonnet 4.6 got within striking distance of Opus 4.6. Same script, different studio. The pressure comes from the competition, and the fight is happening in the middle of the lineup.