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.
The Problem: Security Context Cannot Leave the Network
Closed frontier models run on someone else's infrastructure. Using them sends source code, architecture docs, and unremediated findings outside the network. Aikido points to banks under data-residency mandates and OT operators with no internet route.
Open-weight models solve the residency problem but create a deployment gap. Mixture-of-experts (MoE) models must store every expert, even when a workload uses only a few of them. Security agents also build long-running context. That KV cache competes with model weights for the same GPU memory.
As of 2026, this deployment gap is a central bottleneck for regulated industries: frontier-class security automation must fit within fixed on-prem GPU budgets, and memory efficiency directly determines whether an autonomous agent is viable in an air-gapped SOC.
How Altar-1 Was Built
GLM-5.3 is a 753B parameter MoE model. Each token routes to 8 of 256 experts per layer, which is about 40B active parameters. Aikido applied 2 compression steps:
- Step 1 — Quantization: Altar-1 starts from the cyankiwi GLM-5.3-AWQ-INT4 checkpoint. AWQ stores routed expert weights in 4 bits, with 16-bit activations (W4A16). Attention, the shared expert, dense layers, and the head stay in BF16.
- Step 2 — Pruning: The full GLM-5.3 expert set is reduced to a security-specific subset. The resulting model is 328 GB, down from roughly 1.5 TB, while retaining the security-relevant capabilities needed for autonomous pentesting.
The combination yields a model that fits a single 4x H200 node while keeping the long-context behavior required for security agents that trace multi-step exploits.
via MarkTechPost