← campaign 81ebb2691f9c, execution 1

Comparison: campaign 81ebb2691f9c, execution 1 against campaign bf3a1ac7dff5, execution 1

Newer: campaign 81ebb2691f9c, execution 1, on claude-sonnet-5-5. Earlier: campaign bf3a1ac7dff5, execution 1, on claude-sonnet-5-5. Compared in the order the maintainer chose. Across campaigns only deltas against each execution's own baseline are compared.

Pins that differ

Each arm against its own baseline

CategoryScenarioArmModelMetriccampaign bf3a1ac7dff5, execution 1campaign 81ebb2691f9c, execution 1Change
CS1@1.0baseline-docsclaude-sonnet-5-5M-Q1−1.2 pp n = 3 / 3, range 95.8%–99.4% / 99.4%–99.4% within variance−1.6 pp n = 3 / 3, range 95.8%–99.4% / 99.4%–100.0% within variance−0.4 pp
CS1@1.0baseline-docsclaude-sonnet-5-5M-K1+0.0737 EUR n = 3 / 3, range 0.4034 EUR–0.4343 EUR / 0.3222 EUR–0.3878 EUR beyond variance+0.1154 EUR n = 3 / 3, range 0.4388 EUR–0.4819 EUR / 0.3209 EUR–0.3703 EUR beyond variance+0.0417 EUR
CS1@1.0wingfoilclaude-sonnet-5-5M-Q1−1.2 pp n = 3 / 3, range 95.8%–99.4% / 99.4%–99.4% within variance−0.4 pp n = 3 / 3, range 99.4%–99.4% / 99.4%–100.0% within variance+0.8 pp
CS1@1.0wingfoilclaude-sonnet-5-5M-K1+0.2516 EUR n = 3 / 3, range 0.5927 EUR–0.6152 EUR / 0.3222 EUR–0.3878 EUR beyond variance+0.2402 EUR n = 3 / 3, range 0.5556 EUR–0.6013 EUR / 0.3209 EUR–0.3703 EUR beyond variance−0.0113 EUR
DS2@1.0baseline-docsclaude-sonnet-5-5M-Q1±0.0 pp n = 1 / 1 preliminary±0.0 pp n = 1 / 1 preliminary±0.0 pp
DS2@1.0baseline-docsclaude-sonnet-5-5M-D3±0 n = 1 / 1 preliminary±0 n = 1 / 1 preliminary±0
DS2@1.0wingfoilclaude-sonnet-5-5M-Q1−12.5 pp n = 1 / 1 preliminary±0.0 pp n = 1 / 1 preliminary+12.5 pp
DS2@1.0wingfoilclaude-sonnet-5-5M-D3+3 n = 1 / 1 preliminary±0 n = 1 / 1 preliminary−3
ES8@1.0baseline-docsclaude-sonnet-5-5M-E1−22 n = 1 / 1 preliminary−7 n = 1 / 1 preliminary+15
ES8@1.0wingfoilclaude-sonnet-5-5M-E1−22 n = 1 / 1 preliminary−7 n = 1 / 1 preliminary+15
FS3@1.0baseline-docsclaude-sonnet-5-5M-F1+20.0 pp n = 1 / 1 preliminary+20.0 pp n = 1 / 1 preliminary±0.0 pp
FS3@1.0wingfoilclaude-sonnet-5-5M-F1+20.0 pp n = 1 / 1 preliminary±0.0 pp n = 1 / 1 preliminary−20.0 pp

Not compared

On one side only