DeweyYihengDu

Cache hit rate

96%

Input served from cache

Higher than 69% of developers

Cache reuse

122.7×

Reads per cached token

Higher than 91% of developers

Cost per 1M tokens

$0.65

Blended cost across all tokens

Lower than 77% of developers

Output share

0%

Tokens generated as output

Higher than 59% of developers

Parallel tools

77%

Days with 2+ tools · 2.2 avg/day

Parallel machines

58%

Days with 2+ machines · 1.6 avg, 3 max

Models per active day

3.4

Distinct models used each day

Daily cost variation

1.2

Coefficient of variation; lower is steadier

Active days in span

91%

First to last active day