News
Anthropic prices Claude Opus 5.5 20 percent below its predecessor Opus 5: $4 per million input tokens, $20 per million output tokens, and cache reads cost 60 percent less. OpenAI answered the same day with GPT-6 Sol and Luna, each at half the price of its predecessor – $2 and $10 for Sol, $0.10 and $0.50 for Luna. Cached input tokens get a 90 percent discount.
Our take: applications that read a lot benefit most. Anyone processing long documents or having a second model double-check a result now pays less in production, all the more when the same context is served from the cache again and again. The benchmark and safety figures in both announcements come from the vendors themselves, though. Whether the cheaper model works just as reliably in your own workflow is something only a test on your own data will show.
David Gorges, Managing Director
On 26 and 27 November the Symfony community meets in Warsaw, and five of us are going: Vira, Liubov, Björn, Dmitry Dubinets and David. Three tracks run in parallel, so we split up. Here are the talks we've picked and why.
Mistral released Large 4 as a public preview today. It is a 1.05-trillion-parameter Mixture-of-Experts model with 49B active parameters, text and vision input and a one-million-token context window. Mistral trained it from scratch on 3,800 Nvidia Grace Blackwell GPUs in its own European datacenters and runs it there, end to end, under European law. The weights are due as open weights by the end of October.
It doesn't quite match the leading US models, but if you need your data and your provider to stay in Europe, it is the first serious option at this size.
At DevDay on 29 September 2026, OpenAI announced more than 20 new features across ChatGPT, Codex and the API. The common thread: agents that no longer just answer a question but take on ongoing work, operate computers and react to events in other applications – and, if you want, run entirely inside AWS.
