Performances
RD Popularity Index
RD Growth Index
Comparison
Follow Updates
Follow Lists
People
About GatlingX
GatlingX is building GPU-EVM, a high-performance parallel EVM execution layer. Its core business is to run Ethereum Virtual Machine operations in parallel on GPUs, achieving roughly 100x throughput improvement over traditional CPU-based EVMs. This enables use cases such as training reinforcement learning AI agents for smart contract security auditing, high-speed DeFi simulation and fuzzing, and accelerating Layer 2 networks. The main market pain point it addresses is the sequential execution bottleneck of the standard EVM, which limits transaction throughput and makes large-scale simulation or AI-driven security analysis impractical. By leveraging GPU parallelism, GatlingX aims to make EVM computation dramatically faster and more cost-effective. However, as of August 2026, the project remains in private early access, with no public mainnet or verifiable performance benchmarks released. The official website and Twitter provide limited technical documentation, and no significant development milestones or community adoption have been publicly reported in the past six months. The project's claims of 100x speedup are based on internal benchmarks only, and independent verification is still lacking.
Core progress in the past six months: In March 2026, GatlingX officially released the GPU-EVM source code under a commercial license. The implementation is based on NVIDIA CUDA, using 10,000+ GPU cores to execute the EVM in parallel, claiming 500x throughput gains for parallelizable workloads. At the same time, the team announced it would stop further development of the project and exit crypto infrastructure entirely. Official reasons include: GPU latency is unsuitable for individual transaction processing, declining crypto developer activity, and a polarized market of payments/speculation lacking sustainable business models. Future focus: the core team has pivoted to AI data security, founding a new company called Edison, focused on "deterministic agent runtime security" — limiting an agent's tool-call capabilities after exposure to untrusted content to reduce the blast radius of jailbreaks. The GPU-EVM code remains open, and the team expects external evolution in simulation, payment processing, and security fuzzing. Sources: , , .
Eito Miyamura, Co-founder & CEO of GatlingX. Previously Co-founder & CEO of EdisonWatch (now SealGate), and Co-President of Oxford Blockchain Society. He studied MCompSci Computer Science at University of Oxford (2019-2023) and A-Levels at King's College London Mathematics School. He also interned at Wayve as a Self-driving ML Platform Engineer.