aiqbench takeAI Products

Aikido Security has successfully pruned Z.AI's GLM-5.3 MoE architecture to create Altar-1, a 328 GB open-weight model optimized for on-prem and air-gapped environments. By applying quantization and router-weighted expert pruning without retraining, the release addresses critical data residency concerns for banks and OT operators while maintaining specialized capabilities in coding and reasoning. This development highlights a practical pathway for deploying frontier-grade AI models within restricted network topologies where sending data to external cloud providers is prohibited.

Source coveringMarkTechPost

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Aikido Security Releases Altar-1: An Open-Weight Security Model Pruned From GLM-5.3 to 328 GB

MarkTechPostSeptember 25, 20264 min read

Aikido Security has released Altar-1, its first open-weight security model. It is a compressed version of Z.AI’s GLM-5.3, built to run inside infrastructure the customer controls. Altar-1 powers Aikido Machine, the company’s autonomous pentesting appliance for on-prem and air-gapped networks. Is it deployable? Yes, the weights are public on Hugging Face and run with vLLM […] The post Aikido Security Releases Altar-1: An Open-Weight Security Model Pruned From GLM-5.3 to 328 GB appeared first on MarkTechPost.

Aikido Security has released Altar-1, its first open-weight security model. It is a compressed version of Z.AI’s GLM-5.3, built to run inside infrastructure the customer controls. Altar-1 powers Aikido Machine, the company’s autonomous pentesting appliance for on-prem and air-gapped networks. Is it deployable? Yes, the weights are public on Hugging Face and run with vLLM on a single node of 4x NVIDIA H200 GPUs. Closed frontier models run on someone else’s infrastructure. Using them sends source code, architecture…

Excerpt shown under the publisher's RSS terms · 78 words · full text and figures are on the publisher's site

Read at MarkTechPost ↗Ask about this article