做机器人或边缘部署VLA模型的团队终于有了统一量化方案——内存省71%且性能不降，Pi 0.5和GR00T N1.5用户可以直接用代码复现。
Ω-QVLA是首个无需训练的后训练量化框架，能将视觉-语言-动作模型的语言骨干和扩散动作头统一压缩至W4A4精度，打破此前认为动作头必须混合精度的认知。它通过复合SVD-Hadamard旋转均衡权重能量并分散激活异常值，结合逐步骤的DiT激活缩放量化吸收去噪步间的动态范围漂移。在LIBERO基准上，Pi 0.5和GR00T N1.5经量化后任务成功率分别达98.0%和87.8%，与FP16参考值持平或略优，静态内存占用降低71.3%。真实机器人操作实验也验证了其流畅精准的控制能力。代码已开源。
Ω-QVLA: Robust Quantization for Vision-Language-Action Models via Composite Rotation and Per-step Scaling
Vision-Language-Action (VLA) models unify perception, reasoning, and control within a single policy, yet their multi-billion-parameter backbones and diffusion-based action heads make on-device deployment prohibitively expensive. Prior quantization efforts offer only partial solutions, compressing the LLM backbone while leaving the DiT action head at full precision, or resorting to mixed-precision schemes, driven by the belief that uniformly quantizing the action head is inherently unstable. We challenge this assumption with Omega-QVLA, the first training-free post-training quantization framework that compresses both the language backbone and the entire diffusion action head of a VLA model to a uniform W4A4 precision, eliminating the need for mixed-precision allocation. Omega-QVLA combines a composite SVD-Hadamard rotation that equalizes per-channel weight energy while diffusing residual activation outliers with per-step DiT activation scaling quantization that absorbs dynamic-range drift across denoising steps. On LIBERO, Omega-QVLA compresses Pi 0.5 and GR00T N1.5 to W4A4 with 98.0% and 87.8% task success rates, matching or exceeding their FP16 references of 97.1% and 87.0%, while reducing the static memory footprint by 71.3%. Real-world manipulation experiments further confirm smooth, accurate manipulation where prior methods fail. Code is available at https://github.com/UCMP13753/Omega-QVLA.