{"version":1,"type":"story","url":"https://digestai.news/story/aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3","json":"https://digestai.news/story/aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3.json","markdown":"https://digestai.news/story/aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3.md","slug":"aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3","headline":"Aikido Security releases Altar-1: a 328 GB security model from GLM-5.3","summary":"Aikido Security launched Altar-1, its first open-weight security model derived from Z.AI’s GLM-5.3. The model compresses GLM-5.3’s 753B-parameter mixture-of-experts architecture to 328 GB by pruning 34.4% of experts and quantizing weights to 4 bits. It retains 168 of 256 routed experts per layer, with 8 active per token, and runs on a single node of 4x NVIDIA H200 GPUs using vLLM.\n\nAltar-1 powers Aikido Machine, an autonomous pentesting tool for on-prem and air-gapped networks. Aikido claims it achieves 92% of GLM-5.3’s CVE coverage with a 1-point recall drop, based on an internal benchmark of 32 known vulnerabilities. The model also found a critical-severity vulnerability during a client’s production pentest. Deployment requires Hopper GPUs, and the GLM-5.3 license permits commercial use with a review clause for operators exceeding $10B in annual revenue.","keyPoints":["Aikido Security pruned GLM-5.3 to 328 GB, a 78.2% reduction from its parent model’s size","Altar-1 runs on 4x NVIDIA H200 GPUs with vLLM, supporting air-gapped networks","Recall drops from 65.6% to 60.4% while maintaining 92% of GLM-5.3’s CVE coverage"],"whyItMatters":"Altar-1 lets organizations run large security models on-premises without cloud dependencies, addressing data-residency concerns for banks and OT operators. Its pruning method may help other labs balance size and performance in constrained environments.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["Aikido Security","Z.AI"],"models":["Altar-1","GLM-5.3"],"people":[]},"firstPublishedAt":"2026-09-25T15:14:45Z","updatedAt":"2026-09-25T15:14:45Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"MarkTechPost","title":"Aikido Security Releases Altar-1: An Open-Weight Security Model Pruned From GLM-5.3 to 328 GB","url":"https://marktechpost.com/2026/09/25/aikido-security-releases-altar-1-an-open-weight-security-model-pruned-from-glm-5-3-to-328-gb","publishedAt":"2026-09-25T15:14:45Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Aikido Security releases Altar-1: a 328 GB security model from GLM-5.3\", 25 September 2026, https://digestai.news/story/aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3","publisher":"Digest AI","title":"Aikido Security releases Altar-1: a 328 GB security model from GLM-5.3","datePublished":"2026-09-25T15:14:45Z","url":"https://digestai.news/story/aikido-security-releases-altar-1-a-328-gb-security-model-from-glm-5-3"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}