ENERGY

Energy scales with
spikesnot parameters

Dense nets pay for every weight, every step.
Spiking nets pay when something fires.

Quickstart

pip install thrindex

The model, precisely

E = Σ_layer (pre_spikes × fan_out) × coefficient

Spike events from the pre-synaptic layer

pre_spikes

Post-synaptic connections per neuron

fan_out

Energy per syn-op (default: 0.5 pJ)

coefficient

Spike events from the pre-synaptic layer

Reduction in manual workflow coordination across teams

Energy per syn-op (default: 0.5 pJ)

coefficient

Post-synaptic connections per neuron

fan_out

A real network

A real network

100 timesteps, 10% mean spike rate:

100 timesteps, 10% mean spike rate:

Architecture

Dense(700→512) → LIF → Dense(512→20) → LIF

Architecture

Dense(700→512) → LIF → Dense(512→20) → LIF

Layer 1

7,000 spikes × 512 fan-out = 3,584,000 syn-ops

Layer 1

7,000 spikes × 512 fan-out = 3,584,000 syn-ops

Layer 2

5,120 spikes × 20 fan-out = 102,400 syn-ops

Layer 2

5,120 spikes × 20 fan-out = 102,400 syn-ops

Total

3,686,400 syn-ops × 0.5 pJ ≈ 1.84 µJ per inference

Total

3,686,400 syn-ops × 0.5 pJ ≈ 1.84 µJ per inference

Solutions

Where the math matters

Sparse inputs, always-on ,
tight power budgets, thermal limits.

Sparse inputs, always-on ,
tight power budgets, thermal limits.

Sparse inputs

Event cameras, audio onset, vibration — activity is bursty, not continuous.

Always-on

Network silent when nothing changes. Power collapses toward µW.

Tight power budget

Battery mass is mission mass on drones and orbital payloads.

Thermal limits

Less computation → less heat → smaller cooling on edge nodes.