Safe Rust GPU kernels deliver near‑cuBLAS performance, rivaling vLLM
cuTile Rust lets developers write GPU kernels in safe Rust without losing speed. On NVIDIA B200 it hits 7 TB/s for element‑wise ops and 96% of cuBLAS for GEMM, while its Grout inference engine matches vLLM and SGLang token throughput on large models.