Loom: A Scalable Analytical Neural Computer Architecture
This work provides a novel architecture for neural program execution, enabling scalable and deterministic computation within a transformer loop, which is a step toward bridging neural and classical computing.
Loom introduces a transformer-based computer architecture that executes C programs using fixed, analytically derived weights, achieving program-independent execution with 4.7 million parameters and 928 instruction slots. It can run a 9x9 Sudoku solver in 284 instructions.
We present Loom, a computer architecture that executes programs compiled from C inside a looped transformer whose weights are derived analytically. The architecture implements a 22-opcode instruction set in 8 transformer layers. Each forward pass executes one instruction; the model is applied iteratively until the program counter reaches zero. The full machine state resides in a single tensor $X \in \mathbb{R}^{d \times n}$ of fixed size, and every step has fixed cost for fixed $d$ and $n$, independent of program length or execution history. The default configuration uses $d = 155$ and $n = 1024$, yielding 4.7 million parameters and 928 instruction slots. A compact configuration at $d = 146$ and $n = 512$ suffices for a 9$\times$9 Sudoku solver (284 instructions). The weights are program-independent: programs live in the state tensor, and the same fixed-weight model executes any compiled program. We make Loom source code publicly available at https://github.com/mkturkcan/Loom.