The autonomous multi-model coding agent exits Early Access with enterprise production support and built-in budget governance
Cast AI, the Kubernetes automation platform for cloud-native and AI infrastructure, today announced the general availability of Kimchi Coding, its autonomous multi-model coding agent. Available at kimchi.dev, Kimchi exits Early Access with full production support. Developer teams and enterprises get autonomous AI coding at a fraction of the cost, while ensuring data sovereignty.
Kimchi starts from one premise: the future of AI coding is an orchestration layer that autonomously routes each task to the right model at the right cost, not a single frontier model. Enterprises are already moving this way. They reserve the most capable models for the hardest tasks and let open-weight models handle the bulk of the work. Kimchi’s architecture is built for that workflow.
“We built Kimchi to give every developer frontier-quality AI coding without frontier-sized bills or data risk,” said Laurent Gil, President and Co-Founder of Cast AI. “General availability means that enterprises are now using Kimchi Coding at scale. The economics of AI coding are broken when you rely on a single commercial model. Kimchi fixes that, and does it without asking teams to compromise on quality, security, or control.”
What Kimchi Coding delivers at GA
Kimchi’s core is an autonomous multi-model harness built for token optimization. The orchestration engine routes every task to the best-fit model based on complexity and cost. Feedback loops score the generated code and cut token waste at each step. The cost advantage has two sources: routing that sends most work to open-weight models, and self-hosted inference on GPU infrastructure that Cast AI optimizes down to the node. In shadow-mode evaluations against a commercial-models-only baseline, Kimchi is 2.5x cheaper and matches or exceeds quality on spec-match and test-pass rates.
Budget governance ships in the product, not as an add-on. Hard spend caps apply from individual API keys up to entire organizations. Runaway agentic loops terminate automatically. A real-time FinOps dashboard attributes cost per developer, team, and project.
Kimchi is token-optimized because it runs on Cast AI, the same Kubernetes and GPU optimization engine that made Cast AI the market leader in AI infrastructure. Kimchi is a Cast AI product at its core. That is why it is ISO 27001 and SOC 2 Type II certified from inception, with full GDPR compliance and data sovereignty built in. It runs standalone inside a customer’s own VPC, or on dedicated Nvidia B300 GPUs managed and optimized on Cast AI’s inference infrastructure.
“Kimchi is fundamentally changing how our engineering team thinks about AI-assisted development,” said Dekel Shavit, Senior Director of Engineering at Akamai. “The multi-model approach keeps the quality of a single frontier model while cutting token costs dramatically. And the governance suite is the best we’ve seen in the market. Data sovereignty, cost observability, and budget control in one place. For an organization like Akamai, that makes a real difference.”
About Kimchi Coding
Kimchi is an open-source, terminal-native coding agent and AI assistant designed to streamline agentic software development and AI-assisted coding workflows. Developed by Cast AI, it bridges the gap between AI models and practical engineering, assigning the best-fit model to each step of a task and drawing on frontier models for the hardest work, while open-weight models handle the rest. Kimchi runs serverless on Cast AI’s inference cloud or deploys inside a customer’s own AWS, GCP, Azure, or on-prem environment, with full air-gap support. Learn more and join Early Access at kimchi.dev.
