OpenZL

Lobsters Hottest Tools

Summary

OpenZL is a compression library that generates specialized compressors for specific data formats, achieving high compression ratios at high speeds suitable for datacenter workloads like AI processing.

<p><a href="https://lobste.rs/s/zxt3em/openzl">Comments</a></p>
Original Article
View Cached Full Text

Cached at: 06/27/26, 09:51 AM

# OpenZL Source: [https://openzl.org/](https://openzl.org/) OpenZL delivers high compression ratios*while preserving high speed*, a level of performance that is out of reach for generic compressors\. OpenZL takes a description of your data and builds from it a specialized compressor optimized for your specific format\.[Learn how it works →](https://openzl.org/getting-started/introduction/) OpenZL consists of a core library and tools to generate specialized compressors — all compatible with a single universal decompressor\. It is designed for engineers that deal with large quantities of specialized datasets \(like AI workloads for example\) and require high speed for their processing pipelines\. Here are some examples: - [SAO](https://sun.aei.polsl.pl/~sdeor/corpus/sao.bz2), part of the[Silesia Compression Corpus](https://sun.aei.polsl.pl/~sdeor/index.php?page=silesia): [SAO](https://sun.aei.polsl.pl/~sdeor/corpus/sao.bz2)[zstd](https://github.com/facebook/zstd)\-3xz \-9OpenZLRatiox1\.31x1\.64**x2\.06**Compression Speed115 MB/s3\.1 MB/s**203 MB/s**Decompression Speed890 MB/s30 MB/s**822 MB/s**The result:**much stronger ratios than generic compressors, at the speeds required by datacenter workloads\.** Want to try it yourself? Get started in minutes with the[Quick start guide](https://openzl.org/getting-started/quick-start/)

Similar Articles

@Zai_org: https://x.com/Zai_org/status/2057216685040443743

X AI KOLs Timeline

This paper presents ZCube, a novel network architecture developed by Z.ai, Harnets.AI, and Tsinghua University to address topology-induced congestion in Prefill-Decode disaggregated LLM inference clusters. Production deployments on GLM-5.1 coding workloads achieved a 33% reduction in network CapEx, 15% throughput improvement, and 40.6% reduction in TTFT P99 latency.

@MaxForAI: http://Z.ai and this ZCube paper from Tsinghua—worth a read for anyone in Infra. Many people's first reaction when talking about AI infra is still GPU, memory, quantization, and inference frameworks. But once you get into long context and Prefill-Decode separation, the network is no longer just a 'supporting role' in the data center. Every...

X AI KOLs Timeline

ZCube is a new network architecture that flattens the topology and mixes single/multi-rail access to optimize KV Cache transmission in long-context and PD separation scenarios. In the GLM-5.1 production cluster, it achieved a 33% reduction in switch/optical module costs, a 15% increase in GPU inference throughput, and a 40.6% decrease in TTFT P99.