Metraipolis › archive › dc_d8ec4df38769
SpeedrunBench: Challenging LLM Agents with Video Game Speedrunning — arXiv:2610.08076
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.08076
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:19:22.696Z
- SHA-256 of the archived text
85de0f37fc544774b2b4994a8a1dbefbe93b19314e078c0350e9a2e0ff43ad89- Stable URL of this page
https://metraipolis.com/documento/dc_d8ec4df38769
How to cite
SpeedrunBench: Challenging LLM Agents with Video Game Speedrunning — arXiv:2610.08076 — arXiv (en), https://arxiv.org/abs/2610.08076 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d8ec4df38769 (Metraipolis archive dc_d8ec4df38769, sha256 85de0f37fc544774b2b4994a8a1dbefbe93b19314e078c0350e9a2e0ff43ad89)
BibTeX
@misc{dcd8ec4df38769,
title = {SpeedrunBench: Challenging LLM Agents with Video Game Speedrunning — arXiv:2610.08076},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.08076},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d8ec4df38769; sha256 85de0f37fc544774b2b4994a8a1dbefbe93b19314e078c0350e9a2e0ff43ad89},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08076. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Frontier LLM agents have been shown to be capable of solving increasingly complex tasks for which humans have measurable solutions. This begs the pertinent question of whether LLM agents can go beyond what humans have already solved. The ability to develop sophisticated strategies to tackle consequential problems becomes paramount as well-trodden, human-developed solutions become insufficient for problems for which we lack context or enough training data. We study agents' capability of such strategy formation through the communal practice of video game speedrunning. In speedrunning, practitioners compete to find the fastest way to complete a video game under certain conditions, and in so doing uncovering interesting unorthodox play styles that require a thorough understanding and mastery of the underlying game mechanics. We introduce SPEEDRUNBENCH, a benchmark that evaluates frontier LLM agents across 9 different games. To perform well in this benchmark, agents must repeatedly improve their strategy, reflect on their performance, exploit their gained knowledge, and reason across a long-horizon of actions to improve on an increasingly difficult problem: being faster than themselves and everyone else. Our experiments show that while frontier agents approach human world records in simple platformer games, they remain behind human performance on longer, more complex games under practical budgets. These results suggest that SPEEDRUNBENCH is a useful testbed for studying agents' strategy formation capabilities as well as being a saturation-resistant evaluation measure, as there is almost always a faster completion time waiting to be discovered. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.08076 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T11:15:24.579Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).