TIER IV brings five open E2E driving models into one Autoware portfolio
TIER IV announced five open-source end-to-end autonomous-driving models that can be selected according to camera, LiDAR, radar, and high-definition-map requirements.
AT A GLANCE
Key points
- Five models cover camera-only through LiDAR, radar, and HD-map configurations
- The portfolio spans 16M, 79.9M, and 2B-parameter designs
- Several components still require in-vehicle validation
What is changing?
TIER IV released five end-to-end driving models as open source for Autoware 2.0 on November 11. The portfolio consists of the 2-billion-parameter NVIDIA Alpamayo 2 Edge, the 79.9-million-parameter AutoE2E, the 16-million-parameter METEOR, OnePlanner, and LatentDrive.
The models target different sensor and map requirements. Alpamayo 2 Edge generates reasoning-assisted trajectories from camera input alone. AutoE2E uses a Fast–Slow architecture spanning camera-only setups through LiDAR and radar. METEOR combines map-free multitask perception and planning in one network.
OnePlanner passes LiDAR-based perception features directly into a six-second motion plan, while LatentDrive uses V-JEPA to produce a four-second trajectory from camera video. Some components remain in development or in-vehicle validation, and open-source availability does not by itself demonstrate safety performance.
Explore the original sources
Source date: 11 November 2026.
Published by AI Radar: 11 Nov 2026, 09:08 (Istanbul, UTC+3). Updated: 11 Nov 2026, 10:09.
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…