# Brazilian researchers compare technology stacks for AI agents > A NIC.br and PUC-SP study tested three models, three protocols and four frameworks on 300 Mind2Web tasks. - Canonical URL: https://newsverum.com/en/news/brazilian-researchers-benchmark-agentic-web-technology-stacks - Publisher: Newsverum - Section: Technology - Country: Brazil - Published: 2026-10-12T23:33:38+00:00 - Updated: 2026-10-12T23:33:38+00:00 - Event / data period: 2026-10-09 - Other editions: [Türkçe](https://newsverum.com/tr/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md), [العربية](https://newsverum.com/ar/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md), [Français](https://newsverum.com/fr/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md), [Español](https://newsverum.com/es/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md), [Deutsch](https://newsverum.com/de/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md), [Português](https://newsverum.com/pt/news/brazilian-researchers-benchmark-agentic-web-technology-stacks.md) NIC.br and PUC-SP researchers published a study on 9 October introducing MASTIF, a framework for comparing technology combinations used in multi-agent systems. The evaluation crossed three language models, three communication protocols and four implementation frameworks across 300 tasks from the Mind2Web benchmark. The paper reports that technology combinations differed across measures including task completion, time, concision, tool use and stability of social values; no single stack ranked best on every measure. The authors note limits in the task set and models evaluated. These results provide an experimental baseline, not a guarantee about agents in real-world use. ## Data ### Study scope 300 Mind2Web tasks; MASTIF experiment sample Methodology: Count reported by the paper for its experimental scope; not a national statistic or field result. | Indicator | Experimental scope count | Preprint publication date | |---|---|---| | Tasks evaluated from the Mind2Web benchmark | 300 | 2026-10-09 | Sources: · Download data: [JSON](https://newsverum.com/api/v1/articles/brazilian-researchers-benchmark-agentic-web-technology-stacks/data/study-experimental-scope?lang=en&format=json) · [CSV](https://newsverum.com/api/v1/articles/brazilian-researchers-benchmark-agentic-web-technology-stacks/data/study-experimental-scope?lang=en&format=csv) ## Sources - [arXiv preprint by NIC.br and PUC-SP researchers]() (Source checked: 2026-10-13; Published: 2026-10-09) --- Cite the canonical URL above. [Editorial standards and corrections](https://newsverum.com/en/editorial)