Skip to content

fix(pricing): add display pricing for muse-spark-1.3 and 1.44.0 models - #85

Closed
Star-233 wants to merge 1 commit into
patlux:mainfrom
Star-233:fix/pricing-muse-spark-1.3
Closed

Star-233 wants to merge 1 commit into
patlux:mainfrom
Star-233:fix/pricing-muse-spark-1.3

Conversation

@Star-233

@Star-233 Star-233 commented Sep 5, 2026

Copy link
Copy Markdown
Contributor

Why

Addresses #83 (Bug 1: pricing).

  1. Selecting meta/muse-spark-1.3 or meta/muse-spark-1.3-contributor recorded .00 cost in session metrics because these models were missing from MODEL_COSTS.
  2. Other models added in the 1.44.0 catalog sync (google/gemini-3.8-flash, Qwen/Qwen3.8-Max-0902, meituan/LongCat-2.0:free) were also missing pricing entries.

Cause

In src/pricing.ts, MODEL_COSTS was not updated when the 1.44.0 catalog was merged. index.ts fell back to ZERO_MODEL_COST, silently reporting zero usage cost for unlisted models.

Fix

  • Added pricing entries in src/pricing.ts:
    • meta/muse-spark-1.3: .25 / .25 / .15
    • meta/muse-spark-1.3-contributor: .10 / .20 / .002
    • google/gemini-3.8-flash: .50 / .50 / .15 / .08334
    • Qwen/Qwen3.8-Max-0902: .00 / .00 / .25 / .50
    • meituan/LongCat-2.0:free: .00 (added to freeModels)
  • Regenerated tests/fixtures/commandcode-pricing.json and tests/fixtures/commandcode-model-ids.json.
  • Updated test assertions in tests/test-pricing.ts.

Verification

  • npm run test:pricing — 7 tests pass
  • npm run test:unit — 49 tests pass
  • npm run format:check — clean
  • git diff --check — clean
  • Verified local cost calculation: meta/muse-spark-1.3-contributor correctly calculates non-zero cost matching official rates.

Add pricing entries for `meta/muse-spark-1.3`, `meta/muse-spark-1.3-contributor`,
and other models synced in 1.44.0. Regenerate test fixtures and update
pricing test assertions.
@patlux

patlux commented Sep 15, 2026

Copy link
Copy Markdown
Owner

Thanks for the pricing contribution. Merged #93 now covers these models with refreshed fixtures and explicit regression tests, after full green CI. Reviewed the current official pricing page: Qwen 0902 and Gemini 3.8 cache-write rates are zero, so those proposed values were corrected during consolidation. Closing as superseded, not discarding the fix.

@patlux patlux closed this Sep 15, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants