fix(cloudflare): correct 16 Workers AI catalog prices against live API#102
Open
stackbilt-admin wants to merge 2 commits into
Open
fix(cloudflare): correct 16 Workers AI catalog prices against live API#102stackbilt-admin wants to merge 2 commits into
stackbilt-admin wants to merge 2 commits into
Conversation
…I (2026-07-22 audit) Audited every @cf/ text-generation entry against the account-level /ai/models/search API. 16 entries were wrong: - 12 models listed as free ($0/$0) that are metered — including kimi-k2.6/k2.7-code ($0.95/$4.00) and qwen2.5-coder-32b ($0.66/$1.00), which consumers (aegis cheap-llm/agent-llm ledgers) were reporting as $0 spend. - gpt-oss-120b was 1000x too HIGH ($350/$750 per M — per-M prices had been entered in the per-token fields). - llama-4-scout and llama-3.2-11b-vision had stale rates. All values are USD per token (live per-M price / 1e6). 487 tests pass. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Audited every
@cf/text-generation catalog entry against the live account/ai/models/searchAPI (2026-07-22). 16 entries were wrong:kimi-k2.6/kimi-k2.7-code(actually $0.95/$4.00 per M) andqwen2.5-coder-32b($0.66/$1.00). Consumers computing spend fromusage.cost(e.g. the AEGIScheap_llm_ledger) have been under-reporting Workers AI costs and overstating savings.gpt-oss-120bwas 1000× too high ($350/$750 per M — per-M prices entered in the per-token fields).llama-4-scoutandllama-3.2-11b-visionhad stale rates.All values are USD-per-token derived from live per-M prices. Nothing on Workers AI text generation is free.
Test plan
inputTokenCost/outputTokenCostlines (32 pairs)@cf/models (e.g. glm-4.7-flash run: $0.0002, previously $0.0000)Notes
0aff4c2(current origin tip lineage), not the local WIP checkout.CF_MODEL_PRICING/MODEL_PRICING, shipped v2.38.2) until this publishes — safe to merge and publish in any order.🤖 Generated with Claude Code