What changed: Anthropic released Claude Fable 5.2, with Claude Mythos 5.2 limited to organizations verified under Anthropic's verification programs. Both carry a 1M token context window, 128k max output tokens, and always-on adaptive thinking. Pricing dropped from $10/$50 per MTok (input/output) on Fable 5.1 to $8/$40 on Fable 5.2. The models are available on the Claude API, Claude Platform on AWS, Claude Models on Google Cloud (both Anthropic- and Google-managed), and Claude in Microsoft Foundry. The older computer_20251124 tool is replaced by computer_toolset_20260801, with requests using the old tool now returning a 400 error. A new fast mode for Fable 5.2 delivers up to 2.5x higher output tokens per second at $16/$80 per MTok.
Who is affected
Businesses and developer teams using Claude models via the Claude API, Claude Platform on AWS, Claude Models on Google Cloud, or Claude in Microsoft Foundry.
What it means for your business
For businesses using the Claude API, this means moving from Fable 5.1 to Fable 5.2 lowers input pricing from $10 to $8 per MTok and output pricing from $50 to $40 per MTok, while integrations still calling the old computer_20251124 tool will now fail with a 400 error.
What to do
Migrate any integration still using computer_20251124 to computer_toolset_20260801; update cost projections to the new $8/$40 per MTok pricing; check which direction thinking blocks are preserved or dropped when switching models mid-conversation.
When it takes effect
The source states the models became available as of November 9, 2026; no separate effective date is given.
Source: https://platform.claude.com/docs/en/release-notes/overview#november-9-2026
