A persistent complaint about Claude has been what users call "degradation" in long conversations - sessions where a lot of text has been loaded into the context window (the buffer of information the model can actively reference). Past the 10-prompt mark in heavy sessions, earlier instructions would get fuzzy, responses would shorten, and the model would seem to lose the thread of what was established early on.
Users are now reporting that behavior has noticeably improved. Extended sessions with large amounts of context appear to be holding up better across 10-plus exchanges, with the model maintaining consistency with instructions given at the start of a conversation.
No official announcement from Anthropic accompanied this apparent change. It's unclear whether this reflects a model update, a change in how Claude weights earlier parts of a long conversation, or infrastructure-side improvements. Anthropic periodically updates model behavior without versioning announcements.
The practical test is straightforward: set detailed instructions early in a session, load substantial text, run 15 or more exchanges, and check whether those original instructions are still being respected. For writers, developers, and analysts who run long working sessions in Claude, this change - if it holds up - removes one of the more frustrating parts of using the tool for extended work.