… Interestingly, the Wayback Machine also shows that when the page was updated to include Opus 4.6 and Opus 4.7 in the list, the message remained. It's likely that, while the supported model list was actively updated, the model access information box was not erased and subsequently forgotten. …
… The latest update is GPT-5.5 Instant, and it's the default model for all ChatGPT users. It's paired with an upgraded memory system that shows you which past conversations and saved memories the model pulled from to shape its response. …
… The AI lab highlighted that the model is a "notable improvement on Opus 4.6 in advanced software engineering, with particular gains on the most difficult tasks." Despite being less capable than Mythos the model Anthropic decided is way too dangerous to release publicly right now , Opus 4.7 is still… …
… All of this happens with a single command: /model. You can either browse the available models by executing /model or switch directly by name – /model Opus. …
… They update their models overnight and introduce features you hear about in the news. …
… Claude Code effectively assumes that Anthropic's models will remain the best choice forever, or at least good enough that users won't care about alternatives. But even now, with recent version updates from Google and OpenAI, I am able to see Claude isn’t the only model worth using. …
… A local model is free, and while the hardware required to run this model is on the higher-end, it's possible to offload many of the expert layers to the CPU and still get a model that's capable of generating enough tokens per second to make it useful. …
… In a recent update, Anthropic added a new feature that allows Claude to create interactive visuals for users, enabling the model to generate dynamic charts and diagrams to aid understanding in a way that's never been done before. …
… And perhaps most crucially of all, it was the only model in the test to catch the sneaky wall collision logic on lines 818-819. Anthropic takes the debugging crown Claude's perfect score speaks for itself, but besides that, perhaps what's most telling is how each model approached the problem. …
… The model doesn't learn between sessions, but the journal can so long as you keep it updated. …