modelbenchmark.io

bevformer

NVIDIA · bevformer

Compare

Multi-frame bird's-eye-view 3D perception transformer for autonomous driving

Specification

most-agreed values

Context
128K
Max output
8K
Released
2025-03-18
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
video
Output
text

Price

US dollars per million tokens · most-agreed

Input
free
Output
free

Available from 1 host

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nvidiafreefree——128K8K—

More from NVIDIA

most-hosted first