Scality Launches On-Premises AI Inference Factory to Challenge Cloud Dominance
Event summary
- Scality launched AI Inference Factory on October 8, 2026, an open-code software stack for on-premises AI inference.
- The solution integrates validated open-weight models, disaggregated inference serving, and Scality AI Data Infrastructure (ADI).
- Scality ADI extends key-value cache beyond GPU memory, improving GPU utilization and reducing inference costs.
- The stack supports specialized chatbots, AI-assisted software development, and agentic applications.
- Scality AI Inference Factory is available as a software license or fully managed service.
The big picture
As AI transitions from experimentation to production, enterprises are seeking more predictable costs, greater control over model lifecycles, and increased data sovereignty. Scality's AI Inference Factory positions itself as a viable alternative to cloud-based AI services, addressing these needs with an open-code, on-premises solution. The launch reflects a broader industry trend towards hybrid AI deployments, where critical AI processes are run on-premises while leveraging cloud services for less sensitive workloads.
What we're watching
- Market Adoption
- The pace at which enterprises and government agencies adopt on-premises AI inference solutions to address cost, data sovereignty, and regulatory concerns.
- Technical Performance
- Whether Scality's disaggregated inference serving and extended key-value cache can deliver sustained performance improvements over cloud-based alternatives.
- Competitive Dynamics
- How traditional cloud providers respond to the growing demand for on-premises AI solutions and whether they introduce competing offerings.