docs: add Claude Fable 5.1 to the model choice page - #670
Conversation
Add Claude Fable 5.1 and its five effort variants (claude-5-1-fable-low/medium/high/xhigh/max) to the Anthropic model table in model-choice.mdx, ordered above the existing Claude Fable 5 rows per the newest-first convention. Model IDs and the default-effort variant (high) were verified against warp-server authoritative sources (model/types/ai/types.go, logic/ai/llm/model_choice.go, logic/ai/llm/llm_role.go) and the enabled claude_5_1_fable feature flag. Extend the ZDR data-retention caution below the table to cover Claude Fable 5.1: warp-server's requiresDataRetentionWorkspace and defaultDisabledForEnterpriseLlmIds lists both Fable 5 and Fable 5.1, so the same off-by-default-for-Enterprise / not-available-under-ZDR behavior applies to the new model. Other docs surfaces that mention Fable (pricing-faqs, security-overview, admin-panel, byollm-gemini-enterprise) already use generic provider-retention-requirement language with no version-specific model names, so they already cover Fable 5.1 without changes. Co-Authored-By: Warp <agent@warp.dev>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
@warp-agent-staging[bot] I'm starting a first review of this pull request. You can view the conversation on Warp. I completed the review and no human review was requested for this pull request. Comment Powered by Oz |
|
This PR was generated with Warp. Comment    |
![warp-for-oss[bot]](https://avatars.githubusercontent.com/in/3450139?s=60&v=4){kind=small}
There was a problem hiding this comment.
Overview
This PR adds Claude Fable 5.1 model IDs to the Agent model choice reference and updates the Anthropic data-retention caution to cover the new variants. I did not find correctness, documentation-style, security, or spec-drift concerns in the annotated diff.
Concerns
- None.
Verdict
Found: 0 critical, 0 important, 0 suggestions
Approve
Comment /warp-agent-review on this pull request to retrigger a review (up to 3 times on the same pull request).
Powered by Oz
![warp-factories[bot]](https://avatars.githubusercontent.com/in/2278701?s=60&v=4){kind=small}
There was a problem hiding this comment.
Overview
Small, well-scoped docs change: adds the 5 Claude Fable 5.1 effort variants to the Anthropic model table and extends the ZDR caution to cover Fable 5.1 alongside Fable 5.
Verification
- Model IDs (
claude-5-1-fable-low/medium/high/xhigh/max) match the live model catalog. --changedstyle lint: 0 new issues (the 3 warnings on lines 164/175/191 are pre-existing and outside this diff).- Broken-link check: 0 broken links (3971 internal links scanned).
- Table placement (newest-first, grouped with the existing Fable 5 rows) and caution wording/pluralization are consistent with the rest of the page.
Concerns
- 💡 One inline suggestion below — worth a quick double-check, not blocking.
Verdict
Found: 0 critical, 0 important, 1 suggestion, 0 nits
Approve
| | Claude Opus 5 | `claude-5-opus-xhigh` | Extra high effort | | ||
| | Claude Opus 5 | `claude-5-opus-xhigh-fast` | Fast mode | | ||
| | Claude Opus 5 | `claude-5-opus-max` | Max effort | | ||
| | Claude Fable 5.1 | `claude-5-1-fable-high` | Default effort | |
There was a problem hiding this comment.
💡 [SUGGESTION] The PR description notes there's no explicit "default" flag in the server catalog for Fable 5.1 (or Fable 5) — high is inferred here by matching the Fable 5 row shape. That's a reasonable inference, but since a wrong "Default effort" label would misinform readers about actual out-of-the-box behavior, it's worth a quick confirmation against the client's actual default selection (e.g. the model picker default) before merge, rather than relying solely on precedent-matching with Fable 5.
1 participant
Summary
Adds Claude Fable 5.1 and its five effort variants to the public docs model catalog, per a Slack request from Hongyi Chen.
src/content/docs/agents/inference/model-choice.mdx— addedclaude-5-1-fable-low/medium/high/xhigh/maxto the Anthropic table, ordered above the existing Claude Fable 5 rows (newest-first).highis labeled as the default-effort variant, matching the Fable 5 row shape/precedent (no separate "default" flag exists in the server catalog for either family;highis the most defensible choice given Fable 5.1 has the identical Low/Medium/High/XHigh/Max effort ladder as Fable 5).Verification against warp-server (authoritative catalog source):
model/types/ai/types.go(ClaudeFable51LowEffort...ClaudeFable51MaxEffort) andlogic/ai/llm/model_choice.go(BaseNameClaude51Fable,LLMDisplayNameParts).claude_5_1_fablefeature flag istrueinconfig/prod.yaml, so the model is live in production.logic/ai/llm/anthropic/util/util.go'srequiresDataRetentionWorkspaceandlogic/llm_policy/llm_policy.go'sdefaultDisabledForEnterpriseLlmIdsboth list the Fable 5.1 variants alongside Fable 5, andclient/src/components/admin/utils/uiStrings.ts'sMODEL_ADMIN_TOOLTIPSshows the same Anthropic data-retention tooltip for the Fable 5.1 ids — confirming the same ZDR caution applies.pricing-faqs.mdx,security-overview.mdx,admin-panel.mdx,byollm-gemini-enterprise.mdx) — these already use generic, version-agnostic "provider-specific data retention requirement" / "Claude ... Fable families" language with no hardcoded Fable 5 model names, so they already cover Fable 5.1 without changes. Grepped the rest ofsrc/forclaude-5-fable/claude-5-1-fable/model enumeration surfaces (llms.txt-style listings, changelog, pricing) and found no other place to update.Verification
Docs-only content change (table rows + prose), not user-facing UI code, so no computer-use pass — verified via the repo's documented checks:
npm run build— clean (Complete!, 380 pages generated)python3 .agents/skills/style_lint/style_lint.py --changed— 0 issues on the changed file (3 pre-existing warnings on unrelated lines of the same file, not introduced by this change)python3 .agents/skills/check_for_broken_links/check_links.py --internal-only— 0 broken links across 3971 internal linksOriginating thread: https://warpdev.slack.com/archives/C09BVK0PL3Y/p1788291338237139