Einride’s autonomous trucks to learn from Nvidia’s Alpamayo
A month after agreeing to build its next Einride Driver on Nvidia Hyperion, Einride said it will adapt Alpamayo to its trucks, routes and data, with a road evaluation set for 2027. The post Einride’s autonomous trucks to learn from Nvidia’s Alpamayo appeared first on FreightWaves .
Einride has announced plans to utilize Nvidia's Alpamayo, an AI driving model, within its autonomous driving system. The Swedish firm intends to tailor the model to its own vehicles, routes, and operational data, with a live road evaluation scheduled for the second half of 2027. This adaptation is part of Einride's partnership with Nvidia, which began in September 2023, aimed at developing its next-generation driving system on Nvidia's Hyperion platform.
By employing Alpamayo, Einride anticipates a reduction in development cycles and expenses associated with extending its Einride Driver to new vehicle types and routes. Both companies have not disclosed the financial details of this collaboration. The project aligns with Einride's existing schedule for introducing new vehicle types.
The Swedish company, along with DAF and TNO, is currently defining and testing the integration of Einride's autonomous system into DAF's electric truck platform, with software integration and commissioning set for 2027. As of June 30, 2026, Einride had accumulated over 5,400 hours of driverless operation in contracted customer deployments across the U.S. and Europe.
The company's decision to adopt Alpamayo is linked to its extensive experience in autonomous freight operations for real customers. Nvidia introduced Alpamayo 1 at CES in January, describing it as a 10-billion-parameter vision language action model that generates driving trajectories using reasoning traces. Alpamayo models serve as large-scale teacher models that developers fine-tune and distill into smaller models for their specific driving systems.
In May, Nvidia launched Alpamayo 2 Super, a 34-billion-parameter teacher model designed to be distilled into compact models for in-vehicle use. The specific Alpamayo model that Einride will employ has not been disclosed. Einride plans to post-train and distill Alpamayo using its proprietary operational data. During the shadow mode, the model operates live without controlling the vehicle, with its outputs compared to Einride's driving planner and human driver performance, while existing safety systems remain fully in control.
PlusAI, another company, is also adapting Alpamayo for heavy-duty trucking on Level 4 trucks built on Nvidia's DRIVE AGX Hyperion platform, with fully driverless commercial operations expected to commence in 2027. According to Einride's CTO, Henrik Green, the company's status as the first trucking firm engaged in this endeavor allows it to influence the model's development for freight applications.
Nvidia's Rishi Dhall described the collaboration as an exploration, emphasizing the potential of reasoning-based AI to enhance trucks' navigation in complex driving conditions and expedite the development of new routes and vehicle types.
Written by urgent.news from FreightWaves's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.