feat: recognize Fable/Mythos model families with 1M context window

[coocon]

Problem

Claude Fable 5 / Mythos 5 (claude-fable-5, claude-mythos-5) don't match any built-in model family, so get_context_limit() falls back to the 200k default. These models ship with a 1M-token context window by default — no beta header needed, and the model ID carries no [1m] suffix, so the existing context modifier never triggers either.

Verified against a real Claude Code session: the statusline stdin reports model.id = "claude-fable-5", and the context bar renders used/200k instead of used/1M (e.g. 38% shown when the real usage is 7%).

Reference: Anthropic context windows docs — Fable 5 / Mythos 5 have a 1M window at standard pricing, on by default.

Change

Add two built-in families, reusing the existing version-extraction regex and keeping list order (specific third-party entries and context modifiers still take priority):

BuiltinModelFamily::new("fable", "Fable", 1_000_000),
BuiltinModelFamily::new("mythos", "Mythos", 1_000_000),

Regex behavior (verified with the generated pattern):

model id	display name	context limit
claude-fable-5	Fable 5	1,000,000
claude-fable-5-20260609	Fable 5	1,000,000
claude-fable-5-latest	Fable 5	1,000,000
claude-mythos-5	Mythos 5	1,000,000
claude-sonnet-5	(unchanged)	(unchanged)

Also updated the two comments listing recognized families (models.toml template + Default impl).

Not in scope

Newer Opus/Sonnet versions (Opus 4.6+/5, Sonnet 4.6/5) also moved to 1M-by-default on the API, but the BuiltinModelFamily struct has a single limit per family and older versions of those families are still 200k — making that version-aware is a larger design change, left out of this PR. Fable/Mythos are clean cases: every model in these families is 1M.

🤖 Generated with Claude Code

Summary by Sourcery

Add built-in recognition of Claude Fable and Mythos model families with correct 1M-token context limits.

New Features:

• Recognize Claude Fable and Mythos model families as built-in models with a 1M-token context window.

Documentation:

• Update configuration comments and template text to mention automatic recognition of Fable and Mythos alongside existing Claude families.

[sourcery-ai]

Reviewer's guide (collapsed on small PRs)

Reviewer's Guide

Adds new built-in model families for Anthropic Claude Fable and Mythos models with 1M-token context windows and updates documentation comments describing which Claude families are auto-recognized.

File-Level Changes

Change	Details	Files
Recognize Fable/Mythos Claude model families as built-in with 1M context limits.	Extend the built-in Claude model family list with Fable and Mythos entries using the existing BuiltinModelFamily::new helper and a 1,000,000-token limit. Rely on the existing regex-based version extraction so that claude-fable-5 and claude-mythos-5 variants (including date-suffixed and -latest forms) resolve to these families and get the correct context window.	src/config/models.rs
Update inline documentation to reflect the new auto-recognized families.	Adjust the models.toml template comment to list Fable and Mythos alongside Sonnet, Opus, and Haiku as automatically recognized Claude families. Update the ModelConfig::default comment so it notes that Claude Fable and Mythos models are handled by built-in regex families and do not need explicit entries.	src/config/models.rs

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey - I've left some high level feedback:

• Consider extracting the 1,000,000 context limit into a named constant so that future 1M families (or changes to the default window) can be updated in one place rather than hard-coded per family.
• The new comment "Mythos-class models (Fable 5 / Mythos 5)" may be confusing; consider rephrasing to more directly match Anthropic’s terminology (e.g., "Fable 5 and Mythos 5 models") for clarity.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- Consider extracting the 1,000,000 context limit into a named constant so that future 1M families (or changes to the default window) can be updated in one place rather than hard-coded per family.
- The new comment "Mythos-class models (Fable 5 / Mythos 5)" may be confusing; consider rephrasing to more directly match Anthropic’s terminology (e.g., "Fable 5 and Mythos 5 models") for clarity.

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[coocon]

Addressed the review:

• Comment wording: reworded to "Fable 5 and Mythos 5 models ship with a 1M-token context window by default" (20491db).
• Named constant for 1,000,000: opted to keep the inline literal — the surrounding built-in family list and the Default impl already use inline limits (200_000 ×3, 1_000_000 in the [1m] modifier), so a constant for just these two entries would be inconsistent with local style. Happy to extract one for all limits if the maintainer prefers.