CIQ released Fuzzball 4.2, which connects AI agents directly to GPU infrastructure via native Model Context Protocol (MCP) to secure cluster workflows, health-aware scheduling, and support for AMD ROCm and NVIDIA.
Public source
Publisher name
Public post
Big release from CIQ: Fuzzball 4.2 connects AI agents directly to GPU infrastructure via native Model Context Protocol (MCP). Key value in this update: • AI agents can s…
Company
CIQ
Rocky Linux, Apptainer, Ascender, Warewulf and Fuzzball
- Industry
- Software Development
- Location
- Reno, US
- Company size
- 51–200 employees
About CIQ
CIQ is building the next-generation of secure and performant software infrastructure for the AI era. Founded in 2020 by open source pioneer Gregory Kurtzer, CIQ helps organizations simplify, unify, and democratize high-performance computing and enterprise IT. CIQ is the founding support and services partner of Rocky Linux and the creator of products like Fuzzball, Warewulf Pro, Ascender Pro, and Rocky Linux from CIQ. The company provides scalable, open infrastructure solutions that empower innovation from the operating system up. Rooted in a community-first philosophy, CIQ is trusted by organizations modernizing for a future defined by data and AI. Learn more at https://ciq.com.
See moreLatest activity
Latest activity from CIQ
6 signals
Products & Services
CIQ launched Fuzzball 4.2, which opens the platform to AI agents that can draft, submit, and monitor workflows within limits set by the user.
Research & Knowledge
CIQ CEO Gregory Kurtzer created OpenWALDO, a community-governed training corpus, transparent tooling, and an auditable AI BOM to establish a stronger foundation for open-source AI.
Products & Services
CIQ released Fuzzball 4.2, which ships an MCP server allowing agents to inspect the environment, draft workflows, submit and monitor them, and gate writes and execution behind operator permissions.
Discover more
Similar signals
Similar public activity from other companies.
Products & Services
Red Hat
Red Hat announced that vLLM and llm-d help drive greater GPU utilization and inference efficiency across the hybrid cloud.
Products & Services
Runpod
Runpod noted that AMD Helios is a clear signal that AMD is evolving past the GPU as the product, combining 72 MI455X GPUs, 31 TB of HBM4, EPYC Venice CPUs, and Pensando networking inside an open, liquid-cooled rack with a single-hop UALink-over-Ethernet fabric.
Products & Services
vCluster
vCluster announced the release of Cluster Readiness Engine (NVCRE), a Kubernetes controller that runs real NCCL and training workloads and names every failed node with a reason.
Products & Services
DDN
DDN and NVIDIA power an AI pipeline connecting with ServiceNow to automate ticketing and incident response in live environments using DDN Infinia
Products & Services
Liqid