Skip to content

Getting Started

Installation

pip install aime-loc

pip install aime-loc[eeg,viz]

pip install aime-loc[all]

All Installation Options

Extra Adds Use Case
aime-loc Core SDK AI model scanning
[eeg] MNE, scipy EEG loading, preprocessing
[viz] matplotlib, plotly Charts, figures
[eeg,viz] MNE, scipy, matplotlib Full EEG workflow
[eeg,realtime] MNE, scipy, pylsl Real-time EEG streaming
[mcp] fastmcp MCP server for AI agents
[all] Everything Full toolkit

Get Your API Key

  1. Visit aime-loc.com/signup
  2. Choose your tier:
    • Free: 5 scans/month (great for trying it out)
    • Academic: $99/month, 50 scans (universities, students)
    • Lab: $299/month, 500 scans (research labs)
    • Enterprise: $999/month, 10,000 scans (AI companies)
  3. Copy your API key (format: sk-aime-{tier}_{32chars})

Authentication

export AIME_API_KEY="sk-aime-academic_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6"

from aime_loc import LOC
loc = LOC()  # Reads from AIME_API_KEY

from aime_loc import LOC
loc = LOC(api_key="sk-aime-academic_...")

Path 1: AI Model Analysis

Your First Scan

from aime_loc import LOC

loc = LOC()

# Scan a model (26 questions, ~2 minutes)
profile = loc.scan("meta-llama/Llama-4-Scout")

# View the result
print(profile)
# CognitiveProfile(meta-llama/Llama-4-Scout, TC=14.20%)

print(profile.summary())
# meta-llama/Llama-4-Scout: TC=14.20% (best: Emotion 19.97%, worst: Intuition 11.27%)

# Per-function scores
for func, score in profile.tc_by_function().items():
    print(f"  {func}: {score:.2f}%")

Visualize

profile.radar_chart()                                    # 13-function radar
profile.bar_chart()                                      # Per-function bars
profile.export_figure("profile.png", journal="nature")   # Publication-ready

Compare Two Models

comp = loc.compare("Llama-4-Scout", "DeepSeek-R1")
print(comp.summary())
comp.delta_chart()

Export Data

profile.to_json("profile.json")     # JSON
profile.to_csv("scores.csv")        # CSV
print(profile.to_latex())           # LaTeX table for papers

Next: Scanning Models | Training Audits | Batch Benchmarking

Path 2: EEG Analysis

Your First EEG Score

from aime_loc import LOC
from aime_loc.eeg import EEG

loc = LOC()
eeg = EEG(loc)

# Load any EEG format (auto-detected)
recording = eeg.load("subject01.set")        # EEGLAB
# recording = eeg.load("subject01.edf")      # EDF
# recording = eeg.load("subject01.vhdr")     # BrainVision

# Preprocess (sensible defaults)
recording.preprocess()

# Extract PSD epochs
epochs = recording.extract_epochs(duration=2.0)
print(epochs)
# EpochSet(n_epochs=450, freq_range=0.5-45.0 Hz)

# Score via API (server-side TC scoring)
profile = eeg.score(epochs)
print(profile)
# EEGCognitiveProfile(sub-01, TC=23.40%)

# Visualize
profile.radar_chart()

Consumer Devices

# Load from consumer EEG headset with built-in presets
recording = eeg.load("meditation.csv", device="muse", sfreq=256)
recording.preprocess()
epochs = recording.extract_epochs()
profile = eeg.score(epochs)

Supported devices: muse, openbci_cyton, emotiv_epoc, neurosity, gtec_unicorn

Multi-Subject Study

from pathlib import Path

session = eeg.session()
for f in Path("data/").glob("sub-*/eeg/*.set"):
    rec = eeg.load(f)
    rec.preprocess()
    epochs = rec.extract_epochs()
    session.add(epochs, subject=f.parent.parent.name, task=f.stem)

results = eeg.score_session(session)
results.summary_table()
results.export_csv("study_results.csv")

EEG Visualization

from aime_loc.eeg.viz import psd_plot, timeseries_plot, cognitive_radar

# PSD with mean + standard deviation
psd_plot(epochs, save="psd.png")

# Power time series
timeseries_plot(epochs, profile, save="timeseries.png")

# 13-axis cognitive radar
cognitive_radar(profile, save="eeg_radar.png", journal="nature")

Next: EEG Quick Start | Consumer Devices | Cross-Substrate Comparison

Path 3: Cross-Substrate Comparison

The most powerful feature of AIME LOC — comparing human and AI cognitive profiles using the same framework.

# Score a human EEG recording
human = eeg.score(epochs)

# Score an AI model
llm = loc.scan("meta-llama/Llama-4-Scout")

# Overlay on the same radar chart
from aime_loc.eeg.viz import cognitive_radar
cognitive_radar([human, llm], save="human_vs_ai.png")

Next: Cross-Substrate Comparison

Context Manager

with LOC() as loc:
    profile = loc.scan("meta-llama/Llama-4-Scout")
    profile.radar_chart()
# Connection automatically closed

Async Usage

import asyncio
from aime_loc import AsyncLOC

async def main():
    async with AsyncLOC() as loc:
        profile = await loc.scan("meta-llama/Llama-4-Scout")
        print(profile)

asyncio.run(main())