bevformer

Model release models.dev · Nvidia Source language: Chinese

收录发布日期:2025-03-18(社区数据库,待与官方公告核对)。能力:开放权重;输入:video。上下文:128,000 tokens。定位(来源描述):Multi-frame bird's-eye-view 3D perception transformer for autonomous driving。实际效果:暂无本站独立评测;能力参数不代表评测成绩

Excerpt

收录发布日期:2025-03-18(社区数据库,待与官方公告核对)。能力:开放权重;输入:video。上下文:128,000 tokens。定位(来源描述):Multi-frame bird's-eye-view 3D perception transformer for autonomous driving。实际效果:暂无本站独立评测;能力参数不代表评测成绩

Short excerpt from the source — open the original for the full text.

Sources

← Back to the daily brief