IGNITE Tokamak World Model Architecture
Abstract
We introduce IGNITE, a generative world foundation model for fusion plasma behavior simulation trained in a self-supervised manner from over a decade of unlabeled experimental data at the DIII-D National Fusion Facility. The core of IGNITE is a dynamics model that can simulate DIII-D discharges from a given set of actuator trajectories. These trajectories can be supplied or generated on-the-fly from a textual prompt or from desired experimental outcomes. The model architecture consists of several spatio-temporal tokenizers that embed the different input modalities, including time-series like spatio-temporal measurement data, image sequences, and high-resolution spectrograms, each of which collected at vastly different time scales. The backbone is composed of an auto-regressive dynamics model that has the capacity to predict entire DIII-D discharges given initial latent plasma states and actuator trajectories over a theoretical infinite horizon. IGNITE paves the way towards efficient AI-driven experimental planning and world modeling for nuclear fusion.
Related
- [SUPERSEDED] The Kronos Integrated Platform: A Spherical-Tokamak Breeder and a Deuterium-Helium-3 Burner as One Strategic-Materials and Power Architecture
- SCALE: Simulation-Calibrated Amortized Learning for Energy Materials (A hybrid architecture connecting deterministic modeling, real-world data, and transformer-scale inference for accelerated energy-materials discovery)
- Tokamak-SI: Interpreting Tokamak Discharges with Labeled Catalogs
- Geodesic Acoustic Modes in pair plasmas confined in tokamak magnetic fields
- Mechanism of Ionization Avalanche in Tokamak Microwave Gas Breakdown
BibTeX
@misc{steiner2026ignite,
title = {{IGNITE Tokamak World Model Architecture}},
author = {Peter Steiner and Azarakhsh Jalalvand and Nathaniel Chen and Kouroche Bouchiat and Ricardo Shousha and SangKyeun Kim and Egemen Kolemen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02515},
eprint = {2610.02515},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02515},
}
RIS
TY - UNPB
TI - IGNITE Tokamak World Model Architecture
AU - Peter Steiner
AU - Azarakhsh Jalalvand
AU - Nathaniel Chen
AU - Kouroche Bouchiat
AU - Ricardo Shousha
AU - SangKyeun Kim
AU - Egemen Kolemen
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.02515
UR - https://arxiv.org/abs/2610.02515
AB - We introduce IGNITE, a generative world foundation model for fusion plasma behavior simulation trained in a self-supervised manner from over a decade of unlabeled experimental data at the DIII-D National Fusion Facility. The core of IGNITE is a dynamics model that can simulate DIII-D discharges from a given set of actuator trajectories. These trajectories can be supplied or generated on-the-fly from a textual prompt or from desired experimental outcomes. The model architecture consists of several spatio-temporal tokenizers that embed the different input modalities, including time-series like spatio-temporal measurement data, image sequences, and high-resolution spectrograms, each of which collected at vastly different time scales. The backbone is composed of an auto-regressive dynamics model that has the capacity to predict entire DIII-D discharges given initial latent plasma states and actuator trajectories over a theoretical infinite horizon. IGNITE paves the way towards efficient AI-driven experimental planning and world modeling for nuclear fusion.
ER -