Skip to content

Fix misleading “without pxpipe” cost comparison (#93)#124

Merged
teamchong merged 1 commit into
mainfrom
fix/93-cost-per-request
Jul 19, 2026
Merged

Fix misleading “without pxpipe” cost comparison (#93)#124
teamchong merged 1 commit into
mainfrom
fix/93-cost-per-request

Conversation

@teamchong

Copy link
Copy Markdown
Owner

Fixes #93

The dashboard compared two different kinds of requests.

The current $0.0722 means:

636 total requests − 600 imaged requests = 36 passthrough requests

average cost of those 36 requests

= $0.0722

Those 36 requests were passed through because they were too small to benefit from imaging. The dashboard incorrectly labels their average as “without pxpipe.”
It does not mean the 600 imaged requests would have cost $0.0722 without pxpipe.

Before

600 large requests with pxpipe:       $0.1607/request
36 small passthrough requests:        $0.0722/request

Displayed incorrectly as:

$0.1607 vs $0.0722 without pxpipe

After

600 large requests with pxpipe:       $0.1607/request
Same 600 requests without pxpipe:     $0.4854/request

Displayed correctly as:

$0.1607 vs $0.4854 without pxpipe

The bug is not that $0.0722 was calculated incorrectly. It was labeled and compared incorrectly.

The dashboard labeled the average cost of small passthrough requests as what the larger imaged requests would have cost without pxpipe. Compare each paid imaged request with its own cache-aware text counterfactual instead, and explain the same-request calculation in the math drawer.

Fixes #93
@teamchong
teamchong merged commit c260eae into main Jul 19, 2026
1 check passed
@teamchong
teamchong deleted the fix/93-cost-per-request branch July 19, 2026 02:54
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.

cost per request calc incorrect or...?

1 participant