https://derekwang2002.github.io/ai-intelligence/en/projects/cua-speedrun/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT https://derekwang2002.github.io/ai-intelligence/en/events/ev-20260930-02/ ev-20260930-02-published Thu, 01 Oct 2026 16:01:43 GMT New TRIAL <p>cua-speedrun provides a uniform virtual machine, execution pipeline and agent interface for comparing speed, cost and success across four computer-use benchmarks. The authors find no model family leads all three dimensions; higher reasoning effort can sometimes finish faster, and lower environment latency can increase total task time. Code, infrastructure and analysis are public.</p><p><strong>Why read: </strong>cua-speedrun makes computer-use speed and cost claims cheaply reproducible.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://arxiv.org/abs/2609.40284">arXiv 2609.40284</a> · <a href="https://cuaspeedrun.com">cuaspeedrun.com · Overview</a></p>