Toyota and Woven by Toyota Unveil New AI Technologies to Drive Kakezan
Toyota and Woven by Toyota announced advanced AI technologies designed to accelerate innovation at Woven City.
Woven City AI Vision Engine
A large-scale vision-language model that enables real-time urban understanding and autonomous decision-making.
Technical Specifications
- Model Type: Foundation model for vision-language understanding
- Input: Visual (camera), behavioral (sensor), environmental (telemetry) data streams
- Output: Real-time city response coordination and safety analysis
- Performance: Among world’s leading vision language models on MVBench leaderboard
Capabilities
- Real-time urban scene understanding
- Safety coordination and incident detection
- Multi-modal data integration
- Scalable to additional cities and municipalities
MVBench Performance
- Top-level score on MVBench leaderboard
- Specialized in long-video question answering
- Quantitative validation of VLM capabilities
Deployment & Expansion
Current Implementation
- Proof-of-concept projects: UCC Japan Co., Ltd. partnership
- Location: Woven City (operational since September 2025)
- Use cases: Urban monitoring, infrastructure optimization, safety coordination
Future Plans
- Expansion to additional municipalities
- International partnership opportunities
- Integration with existing Woven City systems
- Behavior AI: Autonomous behavior modeling
- Drive Sync Assist: Autonomous vehicle coordination
Strategic Positioning
Woven City AI Vision Engine demonstrates:
- Toyota’s commitment to intelligent infrastructure
- Practical AI deployment beyond research
- Real-world validation of foundation models
- Integration with automotive and urban systems
Cloud Deployment
Available on AWS Marketplace for broader adoption and testing.
Related Initiatives
Woven City serves as a living laboratory for:
- Autonomous mobility
- Smart city infrastructure
- AI-driven urban management
- Human-AI collaboration in real-world environments