Metraipolis › archive › dc_4a99447ce6ca

TurboMamba: Faster Inference, Faster Training, and Better Prediction with Lexical Memory — arXiv:2610.20682

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20682
License (as archived)
CC0-1.0
Archived
2026-10-16T10:02:07.869Z
SHA-256 of the archived text
36cca8e70e83e1d6d0726b08124c7f0c14acf46ae7894b1e70b62c524bd01182
Stable URL of this page
https://metraipolis.com/documento/dc_4a99447ce6ca

How to cite

TurboMamba: Faster Inference, Faster Training, and Better Prediction with Lexical Memory — arXiv:2610.20682 — arXiv (en), https://arxiv.org/abs/2610.20682 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4a99447ce6ca (Metraipolis archive dc_4a99447ce6ca, sha256 36cca8e70e83e1d6d0726b08124c7f0c14acf46ae7894b1e70b62c524bd01182)
BibTeX
@misc{dc4a99447ce6ca,
  title = {TurboMamba: Faster Inference, Faster Training, and Better Prediction with Lexical Memory — arXiv:2610.20682},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20682},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4a99447ce6ca; sha256 36cca8e70e83e1d6d0726b08124c7f0c14acf46ae7894b1e70b62c524bd01182},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20682. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce TurboMamba, a language-model architecture for faster inference, faster training, and better token prediction. A global Mamba-3 model runs once per completed token block. A small Transformer uses this global context and earlier tokens within the block to predict the next token. Three lightweight additions improve local predictions: direct access to the previous block's last token, a small correction to next-token scores, and a learned mixture with recent-token frequencies. We compare twelve small-scale implementations on public FineWeb-Edu, with comparable parameter counts and shared training-budget caps. The baselines include Transformer, Mamba-3, Jamba, and Zamba. The Jamba and Zamba versions use Mamba-3 in place of their original Mamba modules. TurboMamba achieves the lowest mean test prediction loss (negative log-likelihood, NLL) among all twelve models. Its mean NLL is 9.4% lower than that of Mamba-3, 7.3% lower than that of the Jamba version, and 18.3% lower than that of the Zamba version. End-to-end training throughput is 2.60x that of Mamba-3, 2.81x that of the Jamba version, and 4.82x that of the Zamba version. We measure end-to-end batch latency for a batch of 32 sequences, each with a 256-token prompt and 128 generated tokens. On this workload, TurboMamba achieves a 3.12x speedup over Mamba-3, a 3.53x speedup over the Jamba version, and a 6.21x speedup over the Zamba version.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.20682 — 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 2026-10-16T08:44:47.723Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).