docs: bind routing numbers to the published artifact#35
Merged
Conversation
Contributor
Author
|
CI here is red for an unrelated, pre-existing reason — it is not this PR. This diff touches no Python at all: The Fixed in #36, which is green. Merge that first and this goes green on a rebase. Note that the new |
Three numbers in the routing section were stale and each was maintained by hand: 92% savings against no stated baseline, 14 scoring dimensions, and a free tier described as 9 models. The catalog publishes 8 visible free models, the scorer runs 15 dimensions, and the savings figure is 87% against pinning Claude Opus 5 for every request. They now regenerate from blockrun.ai/brand/numbers.json, checked offline in CI against the committed snapshot. The check is its own job rather than a step, so it does not run three times across the Python matrix. Left alone: the free row still names specific models (GLM-4.7, Llama 4, …). Which models are in the free tier is a catalog question, not a numbers one, and worth its own pass — the count is what this artifact owns.
VickyXAI
force-pushed
the
feat/brand-numbers-sync
branch
from
July 26, 2026 05:20
752c87a to
ab99647
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Consumes blockrun#288's
/brand/numbers.json.What was wrong
The savings figure was an unqualified "up to" against no stated baseline. The replacement is computed from a published workload mix against pinning Claude Opus 5 for every request, so it can be recomputed and checked.
The dimension count was wrong across the whole org: the code comments in ClawRouter said 14, but the scorer builds a 14-entry array and then pushes the agentic dimension — 15 are actually scored. Corrected at the source in ClawRouter#220.
CI
The check is its own job rather than a step in the existing
testjob, so it does not run three times across the Python 3.9 / 3.11 / 3.12 matrix.--checkis offline: it reads the committedbrand-numbers.jsonand never fetches, so a blockrun.ai deploy in progress cannot fail this repo's CI.Left alone deliberately
The free-tier row still names specific models (
GLM-4.7,Llama 4, …). Which models are in the free tier is a catalog question rather than a numbers one — the artifact owns the count, not the roster — and it deserves its own pass.