Put premium-priced models in the $$$$ cost tier - #5786
Open
jurgenwerk wants to merge 1 commit into
Open
Conversation
Lower the top cost-tier bound from a blended $20/M to $15/M so premium-priced frontier models (Claude Fable 5, fast-mode Opus, o3-pro:batch) read as $$$$ instead of sharing $$$ with standard Opus-tier pricing. Also round float representation noise out of the blended cost: $10/$30 blends to 15.000000000000002 in raw float math, which tipped exact-bound prices into the tier above. Rounding to a micro-dollar per million fixes that at every bound (this also corrects two models sitting exactly on the $1 bound from $$ to $). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
jurgenwerk
marked this pull request as ready for review
August 14, 2026 09:04
lukemelia
approved these changes
Aug 16, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This is my suggestion to adjust the model pricing thresholds so that premium priced models are clearly separated from more standard models. For example it was unexpected for me to see Fable 5 in the same pricing bracket as Sonnet 4.6 (labeled with
$$$in the AI assistant)Claude's summary:
The model picker's cost badge maps a blended 3:1 input:output price per million tokens onto fixed dollar-sign bands. The top band started above a blended $20/M, which put premium-priced frontier models — Claude Fable 5 ($10/$50, blended exactly $20) and fast-mode Opus — in the same
$$$band as standard Opus-tier pricing ($5/$25, blended $10) and Sonnet ($3/$15, blended $6). This lowers the top bound to $15/M so those premium models read as$$$$.The blend also picked up binary float noise: $10/$30 blends to 15.000000000000002 in raw float math, tipping a price sitting exactly on an inclusive bound into the band above. The blended value is now rounded to a micro-dollar per million before banding.
Effect on the current catalog:
$$$→$$$$: claude-fable-5, claude-fable-latest, claude-opus-5-fast, claude-opus-4.8-fast, o3-pro:batch$$→$(wobble fix, exactly on the $1 bound): claude-haiku-4.5:batch, seed-2-1-turbo$$$because bounds are inclusive.Catalog cards need a reindex of the openrouter realm to pick up the new computed
costTier/costTierLabelvalues.🤖 Generated with Claude Code