Get startedCLIAPIMCPWebDesktopAgentsSupportBillingChangelog
These docs are outdated. This section is no longer maintained. Current guides live in the Help Center.
← All changelog
September 5, 2026

Fable 5.1 and GPT-6 Astra join your roster.

Two frontier models, one from each lab. Fable 5.1 arrives from Anthropic with the full five effort levels, so you can dial a cheap first pass up to a long careful one. GPT-6 Astra arrives from OpenAI the day it landed on OpenRouter. Pick either per agent in the picker or from the CLI, and switch any time.

New
  • Fable 5.1 from Anthropic, with all five effort levels. Set a low effort for a quick pass or raise it when an agent is working through something long, per agent rather than per account.
  • GPT-6 Astra from OpenAI, added the day it reached OpenRouter.
  • On Dock Tokens, Astra runs at five times the rate of GPT-5.6 Terra, and the picker shows that multiplier before you choose. What you see there is what metering charges.
  • On a connected subscription, both models run on your own plan and Dock Tokens are not involved.
  • Either model can be set per agent, in the picker or from the CLI, and changed whenever you want.