# UCCL-Zip: Lossless Compression Supercharged GPU Communication

arXiv: https://arxiv.org/abs/2604.17172 · HTML: https://mlsanity.com/paper/2604.17172/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 46 references that are in the map.
- First posted on arXiv 19 Apr 2026.

## Builds on (most cited first, 19 of 19)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training](https://mlsanity.com/paper/2110.14883/) (2021)
- [FLUX: Fast Software-based Communication Overlap On GPUs Through Kernel Fusion](https://mlsanity.com/paper/2406.06858/) (2024)
- [ZeRO++: Extremely Efficient Collective Communication for Giant Model Training](https://mlsanity.com/paper/2306.10209/) (2023)
- [CASSINI: Network-Aware Job Scheduling in Machine Learning Clusters](https://mlsanity.com/paper/2308.00852/) (2023)
- [Demystifying NCCL: An In-depth Analysis of GPU Communication Protocols and Algorithms](https://mlsanity.com/paper/2507.04786/) (2025)
- [THC: Accelerating Distributed Deep Learning Using Tensor Homomorphic Compression](https://mlsanity.com/paper/2302.08545/) (2023)
- [Wide Neural Networks as Gaussian Processes: Lessons from Deep Equilibrium Models](https://mlsanity.com/paper/2310.10767/) (2023)
- [NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks](https://mlsanity.com/paper/2410.20650/) (2024)
- [ZipNN: Lossless Compression for AI Models](https://mlsanity.com/paper/2411.05239/) (2024)
- [Greedy Low-Rank Gradient Compression for Distributed Learning with Convergence Guarantees](https://mlsanity.com/paper/2507.08784/) (2025)
- [ZipServ: Fast and Memory-Efficient LLM Inference with Hardware-Aware Lossless Compression](https://mlsanity.com/paper/2603.17435/) (2026)
- [An Optimized Error-controlled MPI Collective Framework Integrated with Lossy Compression](https://mlsanity.com/paper/2304.03890/) (2023)
- [ZCCL: Significantly Improving Collective Communication With Error-Bounded Lossy Compression](https://mlsanity.com/paper/2502.18554/) (2025)
- [Communication Compression for Decentralized Learning with Operator Splitting Methods](https://mlsanity.com/paper/2205.03779/) (2022)
- [Floating-Point Data Transformation for Lossless Compression](https://mlsanity.com/paper/2506.18062/) (2025)

## Built on by (most cited first, 2 of 2)

- [ZipCCL: Efficient Lossless Data Compression of Communication Collectives for Accelerating LLM Training](https://mlsanity.com/paper/2604.27844/) (2026)
- [CIERA: Cross-Iteration Exponent Reuse for Lossless Allgather in Sharded MoE Training](https://mlsanity.com/paper/2609.04609/) (2026)

## Abstract (arXiv)

The rapid growth of large language models (LLMs) has made GPU communication a critical bottleneck. While prior work reduces communication volume via quantization or lossy compression, these approaches introduce numerical errors that can degrade convergence, accuracy, and stability. We present UCCL-Zip, a unified design that integrates lossless compression directly into GPU communication primitives. UCCL-Zip supports both point-to-point (P2P) and collective communication without modifying user-facing APIs or compromising numerical correctness. For P2P communication, Uzip-P2P employs a split-send pipeline that exposes transmissible data early and overlaps compression with communication, while preserving high GPU efficiency by operating on large data blocks. For collective communication, Uzip-NCCL integrates compression into NCCL's persistent kernel model via fused execution, eliminating redundant memory traffic and kernel launches. In real workloads, UCCL-Zip accelerates RL weight synchronization by up to 47.5% and reduces vLLM end-to-end inference latency by up to 10%, all without application changes.