Metraipolis › archive › dc_a24c306fc30c

NPU Hardware Evaluation v1.0 — arXiv:2609.13166

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13166
License (as archived)
CC0-1.0
Archived
2026-10-15T23:17:44.259Z
SHA-256 of the archived text
ab66bef7316430f45ac833789dd5b2800b7bd67dacdb2757853cf8a3e2ecb258
Stable URL of this page
https://metraipolis.com/documento/dc_a24c306fc30c

How to cite

NPU Hardware Evaluation v1.0 — arXiv:2609.13166 — arXiv (en), https://arxiv.org/abs/2609.13166 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a24c306fc30c (Metraipolis archive dc_a24c306fc30c, sha256 ab66bef7316430f45ac833789dd5b2800b7bd67dacdb2757853cf8a3e2ecb258)
BibTeX
@misc{dca24c306fc30c,
  title = {NPU Hardware Evaluation v1.0 — arXiv:2609.13166},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13166},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a24c306fc30c; sha256 ab66bef7316430f45ac833789dd5b2800b7bd67dacdb2757853cf8a3e2ecb258},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13166. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI inference in production settings is becoming the dominant cost line in enterprise AI. The AI inference market is projected to grow from $87B in 2024 to $349B by 2032 (18.9% CAGR). Neural Processing Units (NPUs), chips built specifically for AI inference, are emerging as a compelling alternative to GPU-only architectures, with 35-70% lower power consumption at comparable throughput. This white paper systematically evaluates ten edge AI inference accelerators across three hardware categories: ASIC NPUs (Hailo-8, Hailo-10H, Axelera Metis, Axelera Europa, EdgeCortix Sakura II), SoC DSPs (SiMa MLSoC, Qualcomm QCS6490, QCS8550), and integrated NPUs (Intel Lunar Lake, AMD XDNA2), benchmarked against an NVIDIA RTX A5000 with TensorRT as a production-grade baseline. Twelve reference models spanning convolutional, mobile and transformer architectures are used as a consistent benchmark suite. Results are analysed for throughput, latency, model compatibility, power efficiency, SDK maturity and product lifecycle.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-15 in cs.DC).
Categorie arXiv: Hardware Architecture (cs.AR) ; Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2609.13166 — 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 2026-10-15T22:54:13.003Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).