What each of Ari Wilson's bakes found about coding agents, and what's new at FairBake.https://fairbake.dev/en-ushttps://fairbake.dev/updates/bakeoff-harness/https://fairbake.dev/updates/bakeoff-harness/Running 3 Claude models in Claude Code instead of pi cost 1.8x as much on 10 real tasks. The harness changed the bill, not the fixes.Mon, 09 Nov 2026 00:00:00 GMThttps://fairbake.dev/updates/agent-effort-levels/https://fairbake.dev/updates/agent-effort-levels/Turning coding agents up to high effort roughly doubled how much code they changed, and only 2 of 5 models solved more tasks with it. Fable 5.1's default was its worst setting.Wed, 28 Oct 2026 00:00:00 GMThttps://fairbake.dev/updates/bakeoff-results/https://fairbake.dev/updates/bakeoff-results/277 runs of Claude Code, Codex, and pi on bug fixes from my own repos: an 80x cost spread, speed sorted by harness, and a sandbox setting that hid failures behind green scores.Tue, 22 Sep 2026 00:00:00 GMT