https://electricitybench.com/news Coding-agent scorecards, benchmark results, and provider release coverage from Electricity Bench. en https://electricitybench.com/news/2026-08-05-first-weekly-grading-run https://electricitybench.com/news/2026-08-05-first-weekly-grading-run Wed, 05 Aug 2026 00:00:00 GMT weekly The leaderboard now shows what it was built to show: real agent harnesses, on the subscriptions people actually pay for, graded on the same private task ladder - with the quota each run burned read from the provider's own meter. https://electricitybench.com/news/2026-07-13-hello-electricity-bench https://electricitybench.com/news/2026-07-13-hello-electricity-bench Mon, 13 Jul 2026 00:00:00 GMT announcement Electricity Bench grades new AI models on private, niche task suites and publishes the results here: scorecards, an AI-tools directory, and news. https://electricitybench.com/news/2026-07-12-why-honest-numbers-are-hard https://electricitybench.com/news/2026-07-12-why-honest-numbers-are-hard Sun, 12 Jul 2026 00:00:00 GMT post Publishing a fair number for a model is harder than it looks, and most of the difficulty is invisible in the final figure. https://electricitybench.com/news/2026-07-11-how-we-grade https://electricitybench.com/news/2026-07-11-how-we-grade Sat, 11 Jul 2026 00:00:00 GMT post A grade on Electricity Bench is not a single headline percentage. It is a per-suite roll-up designed to survive contact with reality.