Metraipolis › archive › dc_a969267df109
XStack: A Benchmark for ML Repo Migration — arXiv:2611.08472
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.08472
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T14:42:59.599Z
- SHA-256 of the archived text
ce4ebeb16b70f673f413bde7af8e86931eb7dc21bbb94a6ead6001bf7a66ffae- Stable URL of this page
https://metraipolis.com/documento/dc_a969267df109
How to cite
XStack: A Benchmark for ML Repo Migration — arXiv:2611.08472 — arXiv (en), https://arxiv.org/abs/2611.08472 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a969267df109 (Metraipolis archive dc_a969267df109, sha256 ce4ebeb16b70f673f413bde7af8e86931eb7dc21bbb94a6ead6001bf7a66ffae)
BibTeX
@misc{dca969267df109,
title = {XStack: A Benchmark for ML Repo Migration — arXiv:2611.08472},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.08472},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a969267df109; sha256 ce4ebeb16b70f673f413bde7af8e86931eb7dc21bbb94a6ead6001bf7a66ffae},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08472. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
ML code should run where researchers choose, not where it was first written. Today, machine learning code mainly lives in two ecosystems, PyTorch on NVIDIA GPUs and JAX on Google TPUs. Migrating repositories between them means building and testing across frameworks and hardware. This migration is (1) a practical need, since it unlocks different hardware for researchers, and (2) a natural benchmark: it involves different frameworks and hardware, covers whole repositories, and is long-horizon and challenging. We introduce XStack, a benchmark that evaluates agents on migrating ML repositories between these two ecosystems. Each agent receives a PyTorch repository and a clean TPU machine and must deliver a JAX repository. Its output passes if it meets our grading criteria, based on the source repository's published result, authenticated and calibrated by us. Migration by agents remains unreliable: Claude Opus 5 passes 75%, GPT-5.6-sol 45%, and every other agent at most 40%. XStack is a step toward freeing ML code from its ecosystem. We release XStack to make progress toward reliable migration across frameworks and hardware measurable. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.08472 — 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 2026-11-06T14:18:42.705Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).