feat(anthropic): register claude-opus-4-8 and claude-fable-5 model configs#711
Merged
Merged
Conversation
…nfigs Add the two latest Anthropic models across every model-config surface: - models/configs.py: ModelConfig rows (1M context; 32K first-attempt wire max_tokens with the 128K true cap noted; Opus 4.8 at $5/$25, Fable 5 at $10/$50 per MTok with matching cache rates). Placed after the legacy opus row so prefix fallback for opus-4-1..4-7 is unchanged. - services/pricing.py: new 10/50 tier for Fable 5; exact PRICING and _FAMILY_PREFIXES entries for both models (dated snapshots included). - providers: both added to the anthropic available_models lists; `fable` added to the large-max-tokens pattern so Fable 5 isn't capped at 4096. - query.py: opus-4-8 and fable-5 added to the adaptive-thinking and effort allowlists, fable to the thinking-eligible pattern — without this, opus-4-8 would take the budget_tokens path, a hard 400 on 4.8+. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds the two latest Anthropic models —
claude-opus-4-8($5/$25 per MTok) andclaude-fable-5($10/$50 per MTok), both 1M context / 128K output — to every model-config surface.ModelConfigrows: 1M context window;max_output_tokens=32_000(first-attempt wiremax_tokens; keeps the 64K truncation-escalation meaningful — true cap 128K noted in comments). Rows placed after the legacy opus entry so prefix fallback forclaude-opus-4-1..4-7is byte-for-byte unchanged (test-pinned)._TIER_10_50for Fable 5; exactPRICING+_FAMILY_PREFIXESentries for both, so dated snapshots price correctly.available_modelslists;fableadded to_LARGE_MAX_TOKENS_MODEL_PATTERNso Fable 5 gets the 32K default instead of the legacy 4096.opus-4-8/fable-5added to the adaptive-thinking + effort allowlists andfableto the thinking-eligible pattern. Load-bearing: without it Opus 4.8 takes thebudget_tokenspath, which is a hard 400 on 4.8+ (Fable 5 accepts only adaptive or an omittedthinkingparam).Testing
tests/test_model_system.py— 48 passedclaude-opus-4-7(200K fallback) andclaude-opus-4-1(15/75 tier) unchanged.🤖 Generated with Claude Code