Skip to content

1 million token context for Claude Opus and Sonnet

All Claude models on the 9brains platform now support an extended context window of 1 million tokens. This applies to both Claude Opus 4.6 and Claude Sonnet 4.6.

A token is roughly equivalent to one word. With 1 million tokens you can process about 750,000 words in a single chat, which corresponds to several books or hundreds of business documents at once.

Until now, the context window was limited to 200,000 tokens. With very long documents (e.g. contracts, annual reports, technical documentation) the AI had to cut off parts. With 1M tokens, extensive documents can now be analyzed in full.

When the AI searches your knowledge base, it can now include significantly more relevant passages in its response. This leads to more substantiated and more complete answers, especially for complex questions that combine information from multiple documents.

Even with long conversation histories full of follow-up questions and iterations, the full context is preserved. The AI no longer forgets what you discussed earlier in the chat, even after dozens of messages.

No. The extended context window is used automatically. You can pick Claude Opus or Claude Sonnet via the model selector in the chat as usual. Details on all available models can be found under AI models.