Episode
TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM
- Podcast
- Daily Paper Cast
- Published
- Jul 31, 2026
- Duration seconds
- 1250
- Processing state
not_requested- Canonical source
- https://share.transistor.fm/s/bae73e74
Actions
POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/turbovla-real-time-vision-language-action-model-at-32-hz-on-an-rtx-4090-with-1-gb-vram/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/daily-paper-cast-7079649/turbovla-real-time-vision-language-action-model-at-32-hz-on-an-rtx-4090-with-1-gb-vram.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
🤗 Upvotes: 122 | cs.CV, cs.RO Authors: Hengyi Xie, Chenfei Yao, Xianjin Wu, Xuanyang Xi, Yiping Tang, Di Xu, Yingying Zhu, Dingkang Liang, Xiang Bai, Han Ding Title: TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM Arxiv: http://arxiv.org/abs/2607.27205v1 Abstract: Vision-language-action (VLA) models commonly adopt an LLM-centric $V \to L \to A$ pathway, where visual observations are projected into the representation space of a large language model before being decoded into robot actions. Although effective, this design incurs substantial computation and memory overhead at every policy invocation. In this work, we introduce TurboVLA, a new VLA paradigm that reformulates the conventional $V \to L \to A$ pathway as a direct $V + L \to A$ mapping. Instead of using a large language model as the central interface between perception and action, TurboVLA independently encodes visual observations and language instructions, directly exchanges information between them through lightweight bidirectional vision-language interaction, and predicts continuous action chunks with a compact decoder. This simple design constructs task-conditioned representations directly from visual and linguistic features, significantly reducing the computational and memory costs of VLA inference. On LIBERO, TurboVLA achieves 97.7% average success with only 0.2B parameters, 31.2 ms inference latency, and 0.9 GB inference VRAM on a consumer-grade RTX 4090, matching or outperforming substantially larger VLA policies. These results establish TurboVLA as a simple and effective alternative to the prevailing LLM-centric VLA paradigm, offering a new perspective on how vision, language, and action can be connected for efficient robotic manipulation. Code is available at…