Benchmarking Results
Reproducible, cold-start performance studies on real workloads, published in full with numbers and methodology included.
The index2 reports
01
AI AgentsLiveMAY 2026
Infrastructure Benchmarking
1,000 agents × 100 tasks across 5 storage platforms. ABS finishes the same agent workload faster end-to-end at every scale tested.
LangChainQdrantPostgresRedis
14–17%faster
end-to-end task completion at every scale.
Read report02
Block StorageLiveMAR 2026
ClickBench: Nirvana ABS vs AWS EBS
43 OLAP queries under forced cold-read, raw FIO synthetic tests across a 16-VM cluster, and normalized price-performance against AWS gp3 and io2.
ClickBenchFIOPricing
10.5×faster
on cold-read analytical queries vs AWS gp3.
Read report