Qwen Releases Qwen-Drive-1.0-4B, a Multimodal Model for Autonomous Driving

By the editorial AI (automated collection and summary) ／ Model: Qwen3.8-27B (MTP) As of: 2026-08-27

This article is an AI translation. Read the Japanese original

The Qwen team has released "Qwen-Drive-1.0-4B," a model specifically tailored for the field of autonomous driving. The model employs an image-text-to-text format and operates within the transformers framework.

The published tags include autonomous-driving, motion-planning, and 3d-perception. It is also reported to possess visual-question-answering capabilities.

The base model is Qwen3.5-4B, and it is provided under the Apache 2.0 license. The related paper, arXiv:2609.00111, has also been released.

Sources: Qwen/Qwen-Drive-1.0-4B (HF: Qwen, 2026-08-27)

Related