steve-k

Cache hit rate

96%

Input served from cache

Higher than 51% of developers

Cache reuse

22.3×

Reads per cached token

Higher than 45% of developers

Cost per 1M tokens

$1.56

Blended cost across all tokens

Lower than 16% of developers

Output share

1%

Tokens generated as output

Higher than 89% of developers

Parallel tools

5%

Days with 2+ tools · 1.1 avg/day

Parallel machines

0%

Days with 2+ machines · 1.0 avg, 1 max

Models per active day

2.6

Distinct models used each day

Daily cost variation

0.9

Coefficient of variation; lower is steadier

Active days in span

39%

First to last active day