Unclear Whether these models are available to use is not stated.
What
Claude Code now recognises two new models, Fable 5.2 and Mythos 5.2. Neither was known before.
claude-fable-5-2andclaude-mythos-5-2are new model IDs, with the short namesfable-5-2andmythos-5-2.- The model label shows "5.2" for Fable 5.2. The system prompt, the standing instructions Claude Code sends to the model, now lists Fable 5.2 as the latest Fable model.
- On Google Vertex, a new environment variable,
VERTEX_REGION_CLAUDE_FABLE_5_2, overrides the region used for this model. An environment variable is a named value set in your shell before you start Claude Code. - A set of guidance names appears with the new models:
silent_turn_reminder,original_games_guidance,finishing_work_guidance,memory_guidanceandtimed_silent_turn_reminder.
Fast mode gives faster responses at a higher price. It changes in three ways:
- On Fable 5.2 and Mythos 5.2, fast mode is allowed only if you chose it in this session or the config key
fastModeChosenOnFableOrMythosis true. Both models get their own fast mode pricing entry. Before, fast mode had no such opt-in and fast pricing covered only Opus models. - When fast mode is turned on remotely through
apply_flag_settings, the session keeps its current model if that model supports fast mode. Otherwise it falls back to the default fast model, as it always did before. If the workspace refuses, the error begins "fast mode: workspace did not accept apply_flag_settings". - Turning fast mode off in the settings screen now runs an extra step. Before, it only cleared
fastModeand set the state to off.
Why
If you use Fable 5.2 or Mythos 5.2, you will not end up in fast mode, and paying its higher price, unless you chose it. If your model supports fast mode, turning fast mode on no longer moves you to Opus.
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
[Announcement](https://www.anthropic.com/claude-fable-and-mythos-5-1) · [Migration guide](https://platform.claude.com/docs/en/models/fable-5-2/migration-guide#migrating-from-claude-mythos-5-1-to-claude-mythos-5-2)models/mythos-5-1/overview see the edit
| Claude Mythos 5.2 ([limited availability](https://support.claude.com/en/articles/14604842)) | `anthropic.claude-mythos-5-2` | Yes | Yes | No | No | No |build-with-claude/claude-on-amazon-bedrock-legacy see the edit
On the `bedrock-runtime` endpoint, models that AWS serves through cross-region inference, including Claude Fable 5.2 and Claude Mythos 5.2, take an inference profile ID: the base model ID with a prefix such as `global.` or `us.`, for examp…about-claude/models/model-ids-and-versions see the edit
* **In non-interactive mode**: pass Fable 5.2 as [`--model`](https://code.claude.com/docs/en/cli-reference#cli-flags) and fast mode in [`--settings`](https://code.claude.com/docs/en/cli-reference#cli-flags), for example `claude -p "Summari…fast-mode see the edit
Whether these models are available to use is not stated.
Anthropic's documentation has since written up claude-fable-5-2, on Speed up responses with fast mode.
Added Claude Fable 5.2 (claude-fable-5-2), now the default Fable model — 1M context, $8/$40 per Mtok with $0.20/Mtok cache reads