Anthropic keeps every Claude model firmly under lock. They’ve never released the weights or code to the public. From the original Claude back in 2023 to the cutting-edge Claude Opus 4.6 and Sonnet 4.6 today, everything stays completely closed off. You can only get to these models through claude.ai subscriptions, API access, or their enterprise tools like Claude Code and Cowork. This tight control protects the huge 1-million-token context windows and the models’ leading scores in coding, legal work, finance, and cyber threats.
The approach has shaped language and communication in significant ways. Claude delivers nuanced dialogue that resembles human conversation. It manages sophisticated multilingual interactions with precision, faithfully preserving tone, intent, and cultural subtleties in dozens of languages. Developers depend on it for technical explanations, collaborative drafting, and real-time translation in cross-border teams. Their model bridges linguistic divides, enabling accurate rendering of contracts, medical documentation, and literary texts with little distortion.
Yet the decision to withhold the models restricts wider influence. Open weights would enable adaptation for low-resource languages and offline tools. Instead, Anthropic prioritizes safety by channeling access through controlled channels. Recent attacks from overseas labs highlight the risks: adversaries extract capabilities, strip safeguards, and create unrestricted versions. Strict containment counters these threats.