Effective Tokens replaced by AI Credits
In gh-aw, Effective Tokens (ET) have been replaced by AI Credits (AIC) as the primary spend metric.
This change reflects GitHub Copilot billing and models.dev pricing. It makes spend tracking directly aligned to monetary cost instead of a normalized token proxy.
What this means in practice
Section titled “What this means in practice”gh aw auditandgh aw logsreport AI Credits as the primary spend metric.- Effective Tokens are retired in maintained surfaces and should be treated as historical context.
- Cost reporting and budget discussions should use AIC values.
Verify the change
Section titled “Verify the change”Current gh-aw versions no longer ship the automatic ET-to-AIC fixer.
Update obsolete budget fields in workflow frontmatter by hand,
converting max-effective-tokens to max-ai-credits and
max-daily-effective-tokens to max-daily-ai-credits
(10,000 ET = 1 AIC).
To confirm the migration took effect:
-
Review the rewritten frontmatter:
Terminal window git diff .github/workflowsmax-effective-tokens: 1000000max-ai-credits: 100max-daily-effective-tokens: 5000000max-daily-ai-credits: 500 -
Recompile so the lock files pick up the new limits:
Terminal window gh aw compile
If a workflow still uses max-effective-tokens or
max-daily-effective-tokens, update those values by hand.
Metric reference
Section titled “Metric reference”- AI Credits (AIC): primary spend metric (1 AIC = $0.01 USD)
- Effective Tokens (ET): deprecated legacy metric