Cold Anvil Studios

Cold Anvil Studios — Research2026-08-15T00:00:00Zhttps://coldanvil.com/Jack TylerHarness choice moved a local model 19.4 pointshttps://coldanvil.com/research/jackbench-harness-choice/2026-08-15T00:00:00Z

Qwen 3.6 27B solved 103/144 tasks in Pi and 75/144 in OpenCode on identical tasks with deterministic grading. The hosted baseline (7 models, 1,008 attempts) found one Holm-surviving harness effect.