🧑‍🔬 Tim Dettmers

Affiliations

• Carnegie Mellon University · Pittsburgh, United States — current · 2026 on 4 papers
• Allen Institute for Artificial Intelligence · Seattle, United States — current · 2022–2026 on 3 papers
• University of Washington · Seattle, United States — 2020–2023 on 10 papers
• Università della Svizzera italiana · Lugano, Switzerland — 2015–2018 on 5 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 7 more institutions on single bylines are not listed.

Topics

Efficient models & compression1

Works by year

2020s

Papers

The most cited first (OpenAlex and our library).

1. 8-bit Optimizers via Block-wise Quantization (2021 · 177 citations)
   Stateful optimizers such as Adam and SGD with momentum store per-parameter gradient statistics. These statistics take up a large share of training memory (33–75% by the authors' estimate), which limits how big a model can fit on given hardware. Dettmers et al. show that these statistics can be kept in 8 bits instead…
2. The case for 4-bit precision: k-bit Inference Scaling Laws (2022 · 119 citations)
   Quantizing an LLM shrinks its memory use and, at small batch sizes, its latency. This raises a practical question: for a fixed budget of total model bits, is it better to run a larger model at low precision or a smaller one at higher precision? Dettmers and Zettlemoyer frame this as inference scaling laws that relate…

Also in our library

2 more works of theirs are in our library without a page here yet; they will appear as we publish them.

Co-authors here

3 co-authors with a page, the most papers together first.

• Luke Zettlemoyer ×2
• Mike Lewis
• Sam Shleifer

Cited by

Papers of our library cite their works 455 times (of the 140 884 papers whose bibliographies we read); 21 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/tim-dettmers/