Trendoday

ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters

Frontier Vision-Language Models (VLMs) such as GPT-6 Astra have demonstrated remarkable capabilities in robot control, suggesting that broad pretraining provides knowledge valuable for physical action

  • Upvotes17
–Rank now
#6Peak rank
2Days on list
274Score · 7d
Nov 17, 2026First seen
Rank · 7 days

About this paper

Frontier Vision-Language Models (VLMs) such as GPT-6 Astra have demonstrated remarkable capabilities in robot control, suggesting that broad pretraining provides knowledge valuable for physical action. Yet most Vision-Language-Action (VLA) policies still rely on much smaller VLM backbones, leaving this potential largely untapped. We ask whether frontier-scale representations can provide a stronger foundation for robot learning. We introduce ScaleVLA, which connects a 552B DeepSeek-V4.1-Flash backbone to a trainable action DiT, and systematically compare VLMs from 2B to 552B under a shared training setup. We evaluate three fundamental capabilities of robot foundation models. In sample…

Upvotes
17
Authors
12
Published
Nov 16, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters was first seen on Hugging Face · Daily papers on Nov 17, 2026. Since then it has appeared in 11 checks on 2 different days. Its best position was #6 of 30, reached on Nov 18, 05:28 UTC. It dropped off the list after Nov 18, 05:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 09:2814:2819:2800:2805:28 Nov 17, 09:28 UTC · #15Nov 17, 11:28 UTC · #14Nov 17, 13:28 UTC · #7Nov 17, 15:28 UTC · #7Nov 17, 17:28 UTC · #7Nov 17, 19:28 UTC · #7Nov 17, 21:28 UTC · #7Nov 17, 23:28 UTC · #7Nov 18, 01:28 UTC · #7Nov 18, 03:28 UTC · #7Nov 18, 05:28 UTC · #6

Last 14 days

Best rank per day (UTC).

Nov 9today

Metric history

Platform metric as reported by Hugging Face.

0142843 09:2814:2819:2800:2805:28 Nov 17, 09:28 UTC · 17Nov 17, 11:28 UTC · 19Nov 17, 13:28 UTC · 37Nov 17, 15:28 UTC · 39Nov 17, 17:28 UTC · 41Nov 17, 19:28 UTC · 41Nov 17, 21:28 UTC · 41Nov 17, 23:28 UTC · 41Nov 18, 01:28 UTC · 41Nov 18, 03:28 UTC · 41Nov 18, 05:28 UTC · 43

Questions

Is ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 18, 05:28 UTC, after 2 days of trending.

What is the highest rank ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters reached?

#6 on Hugging Face · Daily papers, on Nov 18, 05:28 UTC. Trendoday checks this list every 120 minutes.

How long has ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters been trending?

It was first seen on Nov 17, 2026 and has been on the list on 2 days (11 checks) since.