SandboxAQ Deploys AI Platform To Accelerate Battery Discovery

SandboxAQ, a Google spinout, released its AQVolt26 research platform to speed discovery of next-generation solid-state battery materials. Backed by $950 million in funding from Alphabet, Nvidia and AI researcher Yann LeCun, the company says AQVolt26 has screened 4,900 halide-based compounds and uses physics-informed AI and LQMs to compress the discovery phase of battery R&D by 90-95%. SandboxAQ is already generating revenue through partnerships and customers including Novonix and the U.S. Army, and plans to monetize via licensing, contract research, and proprietary materials. The platform targets safety, range, and supply-chain resilience by exploring halide solid electrolytes and other solid-state chemistries, but full commercialization remains years away.
What happened
SandboxAQ, a Google spinout, unveiled its updated research platform AQVolt26 to accelerate discovery of solid-state battery materials, claiming it has screened 4,900 halide-based compounds and can cut the discovery phase by 90-95%. The company has raised $950 million from investors including Alphabet, Nvidia, and Yann LeCun, and lists customers such as Novonix and the U.S. Army.
Technical details
AQVolt26 combines physics-forward simulation, machine learning models often called LQMs (Large Quantitative Models), and high-throughput screening to prioritize candidate chemistries. The platform focuses on the early discovery stage, not downstream manufacturing. Key platform capabilities include:
- •High-throughput screening of thousands of halide-based candidates to identify promising solid electrolytes
- •Physics-informed ML to predict ionic conductivity, stability, and interfacial behavior
- •Integration with experimental partners for targeted validation and iterative feedback
Context and significance
Solid-state and halide electrolytes promise better safety and potentially simpler supply chains than conventional liquid-lithium chemistries, which helps address U.S. competitiveness versus China in battery scale and cost. SandboxAQ's approach is to accelerate the most uncertain, time-consuming step of materials R&D so that experimentalists can focus on high-probability leads. The company's revenue model mixes platform licensing, contract research, and proprietary material development, which positions it between pure-play software and materials startups.
Why the claim matters but is cautious
The platform's reported screening and speedups address a real bottleneck: discovery often dominates calendar time. However, SandboxAQ is explicit that AQVolt26 targets phase-one discovery; downstream scale-up, manufacturability, and cell-level performance remain multi-year challenges. As Ang Xiao, who leads SandboxAQ's materials team, said, "We can reduce the time of that [discovery] by 90% to 95%," but later-stage development timelines still apply.
What to watch
Validation by independent experimental partners, detailed performance metrics for candidate electrolytes in full cells, and commercial licensing deals. Watch for demonstrations that a screened candidate advances reliably through scale-up and cell integration to affect cost or supply-chain risk within a practical timeline.
Scoring Rationale
The platform is a meaningful tool for materials discovery with strong funding and early customers, which matters for practitioners in battery R&D and industrial AI. However, the work is discovery-stage, full commercialization and manufacturing impact remain uncertain, and the report is more than three days old, reducing immediate news urgency.
Practice with real Telecom & ISP data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Telecom & ISP problemsStep-by-step roadmaps from zero to job-ready — curated courses, salary data, and the exact learning order that gets you hired.


