Claude Fable 5 and Claude Mythos 5 use the tokenizer introduced with Claude Opus 4.7.

The same text produces roughly 30% more tokens than before Claude Opus 4.7, so prompt measurement matters.

Use the token counting API with model claude-fable-5 to measure prompts under the new tokenizer.

• Jun 11, 2026OpenAI modelcard diff shows the API docs navigation content removed.
• Jun 11, 2026OpenAI pricing page removed visible API pricing content for GPT-5.5, GPT-5.4, and realtime.
• Jun 11, 2026Anthropic launched Claude Fable 5 and Claude Mythos 5 with 1M context and 128k max output.

This is the world we build in. It moves this fast. We move with it.