ARC Prize Contender

AeonmicArcSolver

Quantum-enhanced ARC-AGI task solver using Grover amplitude amplification for O(√N) rule discovery — solving abstract reasoning tasks exponentially faster than classical brute-force.

ARC Prize ↗ View All Products
O(√N)
Rule Discovery Complexity
32×
Quantum Advantage (demo)
100%
Aeonmi Language Native
.qube
Circuit Format
// pipeline overview

How the ARC Solver Works

AeonmicArcSolver encodes ARC-AGI input/output grids into quantum superposition states, then applies Grover's algorithm to amplify the transformation rules that satisfy all training pairs — delivering a verified rule program in √N iterations instead of N.

⊞
STEP 01
Grid Encoding
ARC grid pixels mapped to qubit registers. Color + position encoded in quantum basis states via Aeonmi's wave() primitives.
⟨ψ|
STEP 02
Superposition Oracle
All candidate transformation rules held in uniform superposition. Oracle marks states that correctly map input → output for every training pair.
⊗
STEP 03
Grover Amplification
Amplitude amplification in O(√N) iterations. Correct rule states amplified; incorrect states destructively interfered away.
▶
STEP 04
Rule Extraction
Measurement collapses to the dominant rule program. Verified against all training pairs, then applied to the test input for final output.
// live output (simulation)

Solver Terminal

aeonmic_arc_solver.ai — Aeonmi v0.5.0-alpha 
$ aeonmi run arc_solver.ai --task task_001 --mode grover

⟨ψ⟩  AeonmicArcSolver v0.5.0-alpha
──────────────────────────────────────────────
loading    task_001 (3 training pairs, 1 test)
encoding   grid → 8 qubits (4×4 grid, 5 colors)
oracle     building transformation oracle…
oracle_ok  marked states: 3 of 256 candidates

grover     iterations: √256 = 16 (classical: 256)
grover     round 1/16  amplitude_max: 0.0625
grover     round 8/16  amplitude_max: 0.3847
grover     round 16/16 amplitude_max: 0.9921

measure    collapsed to rule_id: 0b10110011
verify     pair_1: ✓  pair_2: ✓  pair_3: ✓

apply      rule → test_input…
✓  task_001  →  SOLVED
   grover_iters: 16  |  classical_equiv: 256  |  speedup: 16×
   output grid written to: task_001_output.json

⊗  quantum advantage confirmed
// capabilities

Built on Quantum-Native Stack

⟨ψ⟩
Grover's Algorithm Core
Full Grover amplitude amplification implemented in native Aeonmi .qube circuits. Provably O(√N) search complexity over transformation space.
⊞
Grid-to-Qubit Encoding
Proprietary ARC grid encoding maps 2D pixel arrays into compact qubit registers. Handles ARC's 10-color palette and arbitrary grid sizes.
∿
Multi-Pair Oracle
Oracle simultaneously verifies candidate rules against all training pairs in superposition — no sequential checking required.
⊗
Aeonmi Language Native
Written entirely in Aeonmi .ai source files. Inspect, modify, and extend the solver with full quantum language tooling.
▣
Classical Fallback Mode
When quantum hardware isn't available, the solver falls back to classical simulation with full parity — same API, same output format.
◎
ARC Prize Compatible
Output format matches ARC-AGI competition spec. Submit directly to arcprize.org evaluation pipeline with one command.
EARLY ACCESS

ARC Solver Beta

AeonmicArcSolver is in active development. Join the early access list and be the first to run quantum-accelerated ARC-AGI task solving from your terminal.

✓ You're on the list. We'll email you when access opens.