Blog
Insights, updates, and deep dives from the Nirvana Labs team.
Powering AI, blockchain, and
real-time systems.
Powering AI, blockchain, and
Insights, updates, and deep dives from the Nirvana Labs team.
We ran 100,000 LangChain agent tasks across 5 storage platforms. ABS finishes 14-17% faster than io2 at every scale, at 31x less cost.
What LangChain is, how it works, where it hits the cloud, and which part exactly we are benchmarking.
May was about closing the gap between deciding to use Nirvana and actually shipping on it. We launched 26 instance types so teams can pick the compute shape that matches their workload. We opened the Nirvana Examples Library: 18 open-source Terraform templates for running the modern stack. NKS got one-flag auto-scaling. Usage went live so teams can see what's running, what it costs, and what's changed. And we published deep dives on why CPUs matter for agentic AI and what p50/p95/p99 actually m
A LangChain infrastructure benchmark - storage tiers, task completion times, cost-per-task across platforms.
Introducing Usage and Operations: Full lifecycle visibility for every Nirvana resource
The Nirvana Examples library is live
NKS now supports auto-scaling, powered by Karpenter.
Learn what p90, p95, and p99 latency mean
Nirvana Instance Types are live: Compute Families Built for Your Workload
CPUs are back: Why agentic AI needs more CPUs than GPUs, and what's next for cloud
