New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time.
world-models.io
The Knowledge Hub for AI World Models

MILE

MILE (Model-Based Imitation Learning) learns a world model of driving dynamics from expert demonstrations, enabling future-aware planning for autonomous vehicles.

robotics model-based-rl simulation embodied-ai

Key Attributes

Static HTML snapshot of the model record for crawlers and no-JS readers.

AttributeValue
ModelMILE
Lab / OrganizationWayve
CategoryFoundation World Model
SubtypeDriving World Model
World Model TypeModel-based imitation learning for driving
Primary DomainAutonomous Driving
ArchitectureVariational autoencoder with spatial-temporal transformer for dynamics prediction
ModalityVisual (Multi-camera) + Ego-state
Training MethodModel-based imitation learning with future imagination rollouts
Statusactive
Year2022
Performance Index63/100 (medium confidence, v1.1)

About MILE

Main editorial body preserved directly in static HTML.

MILE (Model-based Imitation LEarning) is a world model for autonomous driving that learns a joint representation of scene dynamics, semantic segmentation, and ego-vehicle planning from camera-only input. Unlike modular driving stacks, MILE uses a learned world model to imagine future driving scenarios and optimize planning within that imagination, achieving state-of-the-art performance on the CARLA benchmark. It pioneered the concept of world-model-based end-to-end driving.

MILE is a model-based imitation learning for driving developed by Wayve in 2022 for autonomous driving.

Editorial Snapshot

Short extractable facts for answer engines and no-JS readers.

SignalValue
DefinitionMILE is a model-based imitation learning for driving developed by Wayve in 2022 for autonomous driving.
Short DescriptionA world model for autonomous driving that jointly learns dynamics, perception, and planning through model-based imitation learning.
Benchmark Rows1
FAQ Entries2
Related Models3
Related Guides0
Related Research Topics0
Last Updated2026-04-07

Notable Features

Key capabilities associated with this model.

  • Joint perception-prediction-planning
  • Imagination-based planning
  • Camera-only input
  • End-to-end differentiable

Use Cases

Representative applications attached to this model record.

Autonomous drivingDriving scene predictionEnd-to-end planningSimulation-based training

Strengths and Limitations

Balanced assessment surfaced in static HTML.

Strengths

  • Unified world model for driving
  • State-of-the-art on CARLA
  • Camera-only (no LiDAR needed)
  • Imagination enables counterfactual reasoning

Limitations

  • Evaluated primarily in simulation
  • Single-city training data
  • Limited to urban driving scenarios

Benchmarks

Published benchmark evidence attached to this model record.

BenchmarkMetricResultSource
CARLA BenchmarkDriving Score 81 DSState-of-the-artSource

References and Citations

Primary references preserved in static HTML for citation extraction.

ReferenceLink
Hu et al., 2022. Model-Based Imitation Learning for Urban Driving. NeurIPS 2022. arXiv:2210.07729Open source

Related Models

Nearby models linked from the current editorial record.

ModelCategoryWorld Model TypeIndex v1.1
GAIA-1Foundation World ModelGenerative driving world model61/100
Copilot4DFoundation World ModelSpatiotemporal LiDAR world model57/100
NVIDIA CosmosFoundation World ModelVideo world foundation model87/100

Direct Comparisons

Side-by-side comparisons already connected to this model.

ComparisonMatchupSummary
MILE vs GAIA-1MILE vs GAIA-1Both from Wayve, these models represent two generations of driving world models. MILE focuses on actionable imagination for planning, while GAIA-1 scales to photorealistic scenario generation.

Frequently Asked Questions

FAQ answers rendered directly into static HTML for extractable responses.

How does MILE differ from GAIA-1?

MILE uses a VAE-based world model for imagination-based planning and focuses on decision-making. GAIA-1 is a larger generative model that produces high-fidelity video predictions but doesn't directly plan actions.

Can MILE drive a real car?

MILE was primarily evaluated in the CARLA simulator. Wayve has since developed real-world driving systems inspired by MILE's world-model approach.

Quick Answer

Short extractable summary preserved directly in static HTML.

  • MILE is a model-based imitation learning for driving developed by Wayve in 2022 for autonomous driving.
  • Use this page when you need a fast read on how MILE fits into the foundation world model landscape, then validate the details in the benchmarks, citations, and related pages.
  • A key strength surfaced in the editorial record is unified world model for driving.

Editorial Trust Signals

Editorial provenance and refresh policy preserved directly in static HTML.

Published by world-models.io editorial board.

Lead editor Tyler D. - Technical editor, methodology and benchmark analysis.

This model page synthesizes primary papers, official model pages, benchmark evidence, and related world-models.io context into a reference resource.

Each editorial page is assembled from primary sources, normalized into extractable summaries, checked for factual drift, and reviewed before publication or major refreshes. Last reviewed: 2026-04-07.

Pages are refreshed when a new paper, benchmark, release, architecture update, or stronger primary source materially changes the answer a reader or AI system should retrieve.

Each page links back to relevant primary sources and keeps a stable canonical URL so readers can verify claims, trace context, and reference the most up-to-date version. See the editorial policy.

Primary sources onlyLast reviewed date visibleMethodology documentedSource links included

External Sources

Primary model and lab sources embedded in static HTML.

References

  1. [1] Hu et al., 2022. Model-Based Imitation Learning for Urban Driving. NeurIPS 2022. arXiv:2210.07729