Model catalog is missing Claude Fable 5.1 (claude-fable-5-1) despite carrying the rest of the Claude 5 lineup

[W1ckedS1ck]

Summary

claude-fable-5-1 is absent from the model catalog on the ai-run-sso provider (CodeMie SSO, codemie.lab.epam.com), even though the same catalog already carries the full modern Claude Opus/Sonnet lineage through the current generation (claude-opus-5, claude-sonnet-5). This looks like a gap rather than an intentional exclusion, since Fable is Anthropic's current top-of-line model and every other "latest" Claude tier is already present.

Environment

• CLI: AI/Run CodeMie CLI v0.15.0
• Provider/profile: ai-run-sso, active profile default
• Node: v26.8.1
• OS: macOS (Darwin 25.6.0, arm64)

Reproduction

$ codemie version
AI/Run CodeMie CLI v0.15.0

$ codemie profile
Profile      │ default (Active)
Provider     │ ai-run-sso
Model        │ claude-sonnet-5
CodeMie URL  │ https://codemie.lab.epam.com

$ codemie models list
Provider: ai-run-sso
... 40 models listed ...
$ codemie models list | grep -i fable
(no output)

$ codemie sdk llm list --json | grep -i fable
(no output)

Full Claude entries currently returned by codemie models list / codemie sdk llm list --json (35 total models across all providers, Claude subset below):

claude-4-5-sonnet
claude-4-5-sonnet-vertex
claude-haiku-4-5-20251001
claude-opus-4-5-20251101
claude-opus-4-6-20260205
claude-opus-4-6-vertex
claude-opus-4-7
claude-opus-4-8
claude-opus-5
claude-sonnet-4-5-20250929
claude-sonnet-4-6
claude-sonnet-4-6-vertex
claude-sonnet-5

Notice the Opus track goes 4-5 → 4-6 → 4-7 → 4-8 → 5, and the Sonnet track goes 4-5 → 4-6 → 5, i.e. the catalog is already kept current through the "5" generation for both. Fable is the sibling model at that same generation and is the only one missing.

Why this matters (per Anthropic's official docs)

From https://platform.claude.com/docs/en/models/overview (checked live, current as of today):

"If you're unsure which model to use, start with Claude Opus 5 for most workloads. Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5 at higher effort still fall short."

Side-by-side with what's already supported:

	Claude Fable 5.1 (missing)	Claude Opus 5 (already supported)
Claude API ID	claude-fable-5-1	claude-opus-5
Amazon Bedrock ID	anthropic.claude-fable-5-1	anthropic.claude-opus-5
Google Cloud (Vertex) ID	claude-fable-5-1	claude-opus-5
Microsoft Foundry ID	claude-fable-5-1	claude-opus-5
Use case	Demanding reasoning, long-horizon agentic work	Complex agentic coding, enterprise work
Context window	1M tokens	1M tokens
Max output	128K tokens	128K tokens
Thinking	Adaptive (always on)	Adaptive
Default effort	high	high
Pricing	$10 / input MTok, $50 / output MTok	$5 / input MTok, $25 / output MTok
Reliable knowledge cutoff	Jun 2026	May 2026
Retirement commitment	Not sooner than Sep 1, 2027	Not sooner than Jul 24, 2027

Fable is reachable through the exact same channels CodeMie already sources Claude from (Bedrock / Vertex / Foundry), using the same dateless-ID convention already used for claude-opus-5 and claude-sonnet-5 — so this should be a catalog/config addition on whichever backend feeds ai-run-sso, not a new integration.

Ask

1. Add claude-fable-5-1 to the models surfaced by codemie models list and codemie sdk llm list for the ai-run-sso provider (and any other provider that already lists the rest of the Claude 5 generation).
2. If the model catalog for ai-run-sso is actually owned by a separate internal backend/team and out of scope for this CLI repo, please redirect to the right place — happy to re-file there instead.

Can provide additional CLI debug output (codemie doctor -v) if useful, redacted for credentials.