gpt · 2 arms · 4 effortsgpt-5.2

One model across every effort and arm. Each cell is a run — open it for the full card. Failure chips mark runs with no valid slot.

effort strip

arm \ effort
none
low
high
xhigh
cursor
gpt-5.2 cursor noneopen run ↗
45.9K tokens4 runs · all valid
gpt-5.2 cursor lowopen run ↗
36.6K tokens4 runs · all valid
gpt-5.2 cursor highopen run ↗
41.3K tokens4 runs · all valid
gpt-5.2 cursor xhighopen run ↗
130.1K tokens4 runs · all valid
cursor-xl
gpt-5.2 cursor-xl noneopen run ↗
40.9K tokens4 runs · all valid
gpt-5.2 cursor-xl lowopen run ↗
46.9K tokens4 runs · all valid
no valid slot
no valid slot0 of 4 valid
gpt-5.2 cursor-xl xhighopen run ↗
104.5K tokens4 runs · all valid

effort → output size

Output completion tokens per effort, per arm. A read on how much each model spends as effort rises — not a quality axis.

3.6×low → xhigh output
0100K200Knonelowhighxhigh130.1K36.6K
cursorcursor-xl
gpt-5.2 — model · weather-card-benchmark