Paste your text, or type a token count, and see how much of each Claude model's context window it uses, and whether it fits with room left for the conversation.
0 tokens. Nothing leaves your browser.
Counts use OpenAI's open o200k tokenizer. Claude's own tokenizers usually count somewhat higher, and Opus 4.7+ counts roughly 30% higher, so treat anything above about 70% as tight. For exact Claude counts, use the tokenizer on the home page.
The total number of tokens a model can consider at once: your prompt, the conversation so far, any documents, and the reply it is writing. Text beyond the window is invisible to the model.
Enough for the model's answers and the rest of the conversation. If your document alone uses more than about 70% of the window, expect the model to start forgetting earlier parts of long chats.
They are close but not exact: the page counts with OpenAI's open o200k tokenizer because Claude's tokenizer is not public. Claude typically counts somewhat higher, and Opus 4.7 and later count roughly 30% higher, so leave a margin.
No. Counting happens entirely in your browser. Your text never leaves your device.
MemoryPlugin gives Claude, ChatGPT, Gemini, and 20+ other AI tools one shared long-term memory and a searchable chat history. Tell one AI something once, and the rest have it from your next conversation.
Try MemoryPlugin free