Simple linear attention language models balance the recall-throughput tradeoff

Simran Arora, Sabri Eyuboglu, Michael Zhang, Aman Timalsina, Silas Alberti +3 more

2/12/2026

No official code found. Best implementation: HazyResearch/based (252 stars, Apache-2.0, last commit 2025-06-06).

Semantic Scholar

Code Implementations(1)

HazyResearch/basedCommunity67%

Apache-2.0

Code for exploring Based models from "Simple linear attention language models balance the recall-throughput tradeoff"

25217C++, CudaFeb 7, 20241 years agoApache-2.0

Cite this paper

@article{arora2026simple,
  title  = {Simple linear attention language models balance the recall-throughput tradeoff},
  author = {Simran Arora and Sabri Eyuboglu and Michael Zhang and Aman Timalsina and Silas Alberti and James Zou and Atri Rudra and Christopher Ré},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:a38b4b0c28669ce19e8dcc38e37002c40c9d97a9},
  journal = {ICML 2024 2024}
}

Simple linear attention language models balance the recall-throughput tradeoff

Code Implementations(1)

Cite this paper

Discussion