claude · 2 arms · 5 effortsclaude-opus-4-8

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
low
medium
high
xhigh
max
CC
claude-opus-4-8 CC lowopen run ↗
18.7K tokens4 runs · all valid
claude-opus-4-8 CC mediumopen run ↗
22.8K tokens4 runs · all valid
claude-opus-4-8 CC highopen run ↗
30.1K tokens4 runs · all valid
claude-opus-4-8 CC xhighopen run ↗
55.5K tokens4 runs · all valid
claude-opus-4-8 CC maxopen run ↗
234.8K tokens4 runs · all valid
kiro
claude-opus-4-8 kiro lowopen run ↗
— tokens4 runs · all valid
claude-opus-4-8 kiro mediumopen run ↗
— tokens4 runs · all valid
claude-opus-4-8 kiro highopen run ↗
— tokens4 runs · all valid
claude-opus-4-8 kiro xhighopen run ↗
— tokens4 runs · all valid
claude-opus-4-8 kiro maxopen run ↗
— 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.

12.5×low → max output
0250K500Klowmediumhighxhighmax234.8K18.7K
claude-code (CLI)kiro
claude-opus-4-8 — model · weather-card-benchmark