TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on RTX 4090
H-EmbodVis · hf · 2026-07-30
Existing Vision-Language-Action (VLA) models typically adopt an LLM-centric pathway, projecting visual observations into the language representation space before decoding them into actions. This incurs substantial computation and memory overhead.
To address this, researchers introduced TurboVLA, a new paradigm that reformulates the conventional pathway into a direct "Vision + Language to Action" mapping. Instead of using an LLM as the central interface, TurboVLA independently encodes visual and linguistic inputs, exchanges information via lightweight bidirectional interaction, and predicts continuous action chunks with a compact decoder.
On the LIBERO benchmark, TurboVLA achieves a 97.7% average success rate using only 0.2B parameters. On a consumer-grade RTX 4090, it delivers a 31.2 ms inference latency (32 Hz) and requires less than 1 GB of VRAM, matching or outperforming significantly larger VLA policies.
Related event: TurboVLA Enables 32Hz Robot Control on Consumer GPUs(2 posts)→
More from Embodied
- Autonomous vehicle demo shows car reversing to break narrow-road deadlock, with a robotic arm to shift gears — sanjeevs_iitr · 2026-10-16
- Surface Laptop Ultra review: a beautiful $2,599 laptop for an AI future that hasn't arrived — tomwarren · 2026-10-16
- Waymo brings robotaxis to Toronto, targeting rider service in 2028 — CSProfKGD · 2026-10-16
- Musk shows Cybertruck hovering on jets, delivering on 'flying cars' promise — XFreeze · 2026-10-16
- Unitree Turns Spring Festival Gala Performance into UniUni Robot Swarm Choreography Product — CyberRobooo · 2026-10-16
- Tesla FSD (Supervised) approved in Slovakia, the 9th EU country — XFreeze · 2026-10-16