← back to paper

Review history

arxiv: 2607.27205 · 2 revisions

TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM

1. 2026-07-30 CONDITIONAL HIGH v1.2.0-grok45-kimik3 novelty 5.5
   79787 ms 22708 in 3807 out 2026-07-30T10:54:04.146928+00:00
2. 2026-07-30 CONDITIONAL LOW v1.1.0-grok45 novelty 5.5
   48980 ms 6218 in 2470 out 2026-07-30T09:42:24.246020+00:00