Open AI Research Lab

Building intelligence that is open, accountable, and shared.

Quanta is an independent research lab developing frontier models and releasing them openly — weights, training recipes, and evaluations — so the broader community can study, audit, and improve them.

38
Papers published
14
Open model releases
2.1M
Model downloads
100%
Permissively licensed
Our Mission
Research in the open, by default.

We believe the most capable models should not be locked behind closed doors. Transparency is not a constraint on progress — it is the mechanism by which progress becomes trustworthy.

Every system we build ships with its weights, data documentation, training code, and a full evaluation suite. We publish negative results alongside the positive ones, because reproducibility is a feature, not an afterthought. Our work spans pre-training efficiency, interpretability, and alignment — connected by a single commitment: the people who use these systems deserve to understand them.

Open weights Reproducible training Independent audits Permissive licensing
Research Directions

Three problems we think are foundational.

Each direction is led by a small, focused team and judged by what it contributes back to the open ecosystem.

Efficient Pre-training

Scaling laws beyond raw compute. We study data curation, sparse architectures, and curriculum design to train frontier-quality models at a fraction of the cost — and we release the recipes.

View work

Interpretability

Reverse-engineering the internal computations of large models. We develop tools to locate circuits, trace features, and explain behaviour — turning black boxes into objects of study.

View work

Alignment & Safety

Methods for making model behaviour predictable, controllable, and robust to misuse. We build open evaluations and red-teaming protocols that anyone can run and extend.

View work
Recent Papers

From the lab, recently.

Full archive →
Mar 2025
Pre-training

Quanta-7B: An Open Recipe for Sub-$100k Frontier Models

A. Reyes, M. Okonkwo, L. Tanaka, and the Quanta Pre-training Team
Read PDF
Feb 2025
Interpretability

Tracing Refusal Circuits Across Model Scales

S. Bauer, N. Costa, and J. Whitfield
Read PDF
Jan 2025
Alignment

OpenEval: A Reproducible Benchmark Suite for Open Models

P. Andersson, R. Mehta, and the Safety Group
Read PDF
Dec 2024
Pre-training

Data Curation as a Scaling Lever: Lessons from 2 Trillion Tokens

L. Tanaka, K. Brennan, and A. Reyes
Read PDF
Nov 2024
Interpretability

Sparse Autoencoders Recover Interpretable Features at Scale

N. Costa, S. Bauer, and D. Park
Read PDF
Careers

Do the work you'll want to publish.

We hire researchers and engineers who care about getting the details right and sharing what they learn. Small teams, long horizons, real ownership — and everything you build sees the light of day.

See open roles →
Research Scientist — Pre-training ML Engineer — Infrastructure Research Engineer — Interpretability Safety Researcher