Arena Physica releases Heaviside-1 foundation model for electromagnetism
seanmcdonaldxyz · x · 2026-09-02
Arena Physica has unveiled Heaviside-1, a second-generation foundation model for electromagnetism aimed at building electromagnetic superintelligence.
Key Updates
- Scale & Data: The model has 350M parameters (roughly GPT-2 size), 10x larger than its predecessor, trained on 250k unique designs with >500B unique EM field samples.
- Capabilities: It natively encodes 3D structures, material properties, and excitation patterns as input, predicting full EM fields at arbitrary locations.
- Performance: It runs 100,000x faster than commercial solvers while maintaining accuracy within 1 dB. Field supervision enables strong out-of-distribution generalization, reducing S-parameter error from 0.99 dB to 0.53 dB.
The model is available in beta via Atlas Fields Studio.
Related event: Arena Physica releases Heaviside-1 electromagnetic foundation model(2 posts)→
More from Research
- TRACE Open-Sourced: Migrating G1 Controller to Full-Size Humanoid H2 in 8 Hours on One GPU — zhengyiluo · 2026-10-18
- Humanoid Robot Funding Tops $3B as Brian Potter Explains the VLA Models Driving Robots — scottleibrand · 2026-10-18
- Terence Tao on 'War Footing' as OpenAI Claims Navier-Stokes Solve Shakes Mathematics — AlexTensor · 2026-10-18
- Univ. of Tokyo paper: why your recorded voice feels eerie — a vocal uncanny valley — ssh4net · 2026-10-18
- MIT's SANDO Drone Planner Avoids Every Moving Obstacle Across 12 Real Flights — Olivier__OG · 2026-10-18
- No-backprop pretraining speedup of 2.3x claimed, inspired by mech interp essay — CatAstro_Piyush · 2026-10-18