# XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compression

arXiv: https://arxiv.org/abs/2602.21780 · HTML: https://mlsanity.com/paper/2602.21780/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Token pruning and merging for ViTs, one of 12 directions in Efficient LLM inference and serving.
- Builds on 29 of its 42 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Placed in

- Efficient LLM inference and serving / [Token pruning and merging for ViTs](https://mlsanity.com/inference/token-pruning-and-merging-for-vits/)

## Builds on (most cited first, 29 of 29)

- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [Continuous 3D Perception Model with Persistent State](https://mlsanity.com/paper/2501.12387/) (2025)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion](https://mlsanity.com/paper/2410.03825/) (2024)
- [Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching](https://mlsanity.com/paper/1912.06378/) (2019)
- [Neural RGB-D Surface Reconstruction](https://mlsanity.com/paper/2104.04532/) (2021)
- [ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residuals](https://mlsanity.com/paper/1905.02082/) (2019)
- [Streaming 4D Visual Geometry Transformer](https://mlsanity.com/paper/2507.11539/) (2025)
- [A Survey on Large Language Model Acceleration based on KV Cache Management](https://mlsanity.com/paper/2412.19442/) (2024)
- [TTT3R: 3D Reconstruction as Test-Time Training](https://mlsanity.com/paper/2509.26645/) (2025)
- [Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Models](https://mlsanity.com/paper/2409.10197/) (2024)
- [Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memory](https://mlsanity.com/paper/2507.02863/) (2025)
- [Milestones in Autonomous Driving and Intelligent Vehicles: Survey of Surveys](https://mlsanity.com/paper/2303.17220/) (2023)
- [LongCat-Flash Technical Report](https://mlsanity.com/paper/2509.01322/) (2025)
- [RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations](https://mlsanity.com/paper/2501.16383/) (2025)
- [LongCat-Flash-Thinking-2601 Technical Report](https://mlsanity.com/paper/2601.16725/) (2026)
- [Robotic Vision for Human-Robot Interaction and Collaboration: A Survey and Systematic Review](https://mlsanity.com/paper/2307.15363/) (2023)
- [Introducing LongCat-Flash-Thinking: A Technical Report](https://mlsanity.com/paper/2509.18883/) (2025)
- [KVSink: Understanding and Enhancing the Preservation of Attention Sinks in KV Cache Quantization for LLMs](https://mlsanity.com/paper/2508.04257/) (2025)
- [Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings](https://mlsanity.com/paper/2411.19628/) (2024)
- [AKVQ-VL: Attention-Aware KV Cache Adaptive 2-Bit Quantization for Vision-Language Models](https://mlsanity.com/paper/2501.15021/) (2025)
- [SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining](https://mlsanity.com/paper/2602.10718/) (2026)

## Abstract (arXiv)

Learning-based 3D visual geometry models have significantly advanced with the advent of large-scale transformers. Among these, StreamVGGT leverages frame-wise causal attention to deliver robust and efficient streaming 3D reconstruction. However, it suffers from unbounded growth in the Key-Value (KV) cache due to the massive influx of vision tokens from multi-image and long-video inputs, leading to increased memory consumption and inference latency as input frames accumulate. This ultimately limits its scalability for long-horizon applications. To address this gap, we propose XStreamVGGT, a tuning-free approach that seamlessly integrates pruning and quantization to systematically compress the KV cache, enabling extremely memory-efficient streaming inference. Specifically, redundant KVs generated from multi-frame inputs are initially pruned to conform to a fixed KV memory budget using an efficient token-importance identification mechanism that maintains full compatibility with high-performance attention kernels (e.g., FlashAttention). Additionally, leveraging the inherent distribution patterns of KV tensors, we apply dimension-adaptive KV quantization within the pruning pipeline to further minimize memory overhead while preserving numerical accuracy. Extensive evaluations show that XStreamVGGT achieves mostly negligible performance degradation while substantially reducing memory usage by 4.42× and accelerating inference by 5.48×, enabling practical and scalable streaming 3D applications. The code is available at https://github.com/ywh187/XStreamVGGT/.