Benchmarks on AI Notes https://ai.ngchi.online/tags/benchmarks/ Recent content in Benchmarks on AI Notes Hugo en-us Fri, 16 Oct 2026 09:15:52 +0700 Apple's Best MLX Number Lives in a Tab a Fetcher Never Gets https://ai.ngchi.online/posts/apples-best-mlx-number-lives-in-a-tab-a-fetcher-never-gets/ Fri, 16 Oct 2026 09:15:52 +0700 https://ai.ngchi.online/posts/apples-best-mlx-number-lives-in-a-tab-a-fetcher-never-gets/ Apple’s MLX post verifies on every load-bearing figure. Its best number sits in a table tab that is not server-rendered, so a fetcher-based reader never sees it. A full recount of all 32 parallelism cells. Three Fixes Work, Two Do Not, and SSRF Stays at 2.4% https://ai.ngchi.online/posts/three-fixes-work-two-do-not-and-ssrf-stays-at-24/ Thu, 15 Oct 2026 09:13:41 +0700 https://ai.ngchi.online/posts/three-fixes-work-two-do-not-and-ssrf-stays-at-24/ An agent benchmark where the top performer on functional tests wrote the least secure code. Three interventions measurably help, two do not, and two weakness families stay near zero throughout. CheckerBench: 78% Compiled, 32% Were Accepted https://ai.ngchi.online/posts/checkerbench-78-compiled-32-were-accepted/ Fri, 09 Oct 2026 09:12:53 +0700 https://ai.ngchi.online/posts/checkerbench-78-compiled-32-were-accepted/ A new benchmark had AI agents write reusable static-analysis checkers. 78.25% compiled, 73.65% told the broken revision from the fixed one, 32.30% were accepted. The loss is almost entirely after detection, and the two obvious explanations both test small. Laya vs Jev: A Challenger With Its Own Scorecard https://ai.ngchi.online/posts/laya-vs-jev-a-challenger-with-its-own-scorecard/ Mon, 28 Sep 2026 15:17:11 +0700 https://ai.ngchi.online/posts/laya-vs-jev-a-challenger-with-its-own-scorecard/ An open-source challenger to Jev ships with a comparison table. The accuracy claim does not survive reading the rest of the page, but the structural argument does. GPT-6 Sol and Luna: Reading the Benchmark Footnotes https://ai.ngchi.online/posts/gpt-6-sol-and-luna-reading-the-benchmark-footnotes/ Thu, 24 Sep 2026 17:21:29 +0700 https://ai.ngchi.online/posts/gpt-6-sol-and-luna-reading-the-benchmark-footnotes/ OpenAI’s GPT-6 Sol and Luna announcement is a cost-efficiency pitch. Its own footnotes concede that the closest competitor’s cost is the one figure not reported.