Skip to content
Agientry
← Back to Videos

VFlash Turns Visual Planning Into Parallel Decoding

Loading video
Loading video

VFlash converts sequential visual-plan decoding for boxes, points and trajectories into a parallel process, giving physical-AI VLM models faster reactive capabilities.

Physical AIVLMVisual PlanningParallel DecodingNVIDIA
Category: research
Author: @ZhidingYu
Date: 2026-11-05T00:00:00
Duration: 19.333s
Source: https://x.com/ZhidingYu/status/2118399296693498254
Reference: https://research.nvidia.com/labs/lpr/vflash/

Related Videos

VFlash Gives Physical AI a System-1 Visual Planner
0:56

VFlash Gives Physical AI a System-1 Visual Planner

@ZhidingYu · 5 views · 2026-11-05
Physical AIVLM
Walden Robotics raises 300M at 1.1B valuation
0:15

Walden Robotics raises 300M at 1.1B valuation

@LeoKharon · 454 views · 2026-07-26
Walden RoboticsFunding
Hyundai Partners NVIDIA on Robot Reference Platform
1:08

Hyundai Partners NVIDIA on Robot Reference Platform

@clankrmedia · 473 views · 2026-07-25
NVIDIAHyundai
D&A Turns Any VLM Into a Robot Reward Model
0:33

D&A Turns Any VLM Into a Robot Reward Model

@memmelma · 44 views · 2026-11-03
Robot RewardsVLM
PixelUMM: Pixel-Space Image and Video Model
0:29

PixelUMM: Pixel-Space Image and Video Model

@CongWei1230 · 85 views · 2026-10-01
NVIDIAPixelUMM
Nemotron 3.5 Transcribe: One Model for Multilingual Audio
2:00

Nemotron 3.5 Transcribe: One Model for Multilingual Audio

@NVIDIAAI · 98 views · 2026-10-28
NVIDIANemotron

As an Amazon Associate, we earn from qualifying purchases.