Infino benchmarks
Every number links to the run behind it and a way to reproduce it. Where another engine is faster, the page says so.
Pick a benchmark
- ClickBench SQL analytics over 100M rows, on the public board open →
- Search on object storage Full-text, vector and hybrid latency, measured in CI open →
- VectorDBBench Vector databases at Cohere 1M and 10M vectors open →
- Search Benchmark, the Game Full-text at Wikipedia scale, with Tantivy and Lucene open →
- Embedded vector libraries In one process with turbovec and FAISS open →
Common questions
Can I reproduce these numbers?
Yes. ClickBench runs from the infino directory in the upstream ClickBench repository. The search latencies run with cargo bench in the engine repository. VectorDBBench, the full-text benchmark and the vector library comparison run from public repositories, and each published result records the engine commit and the command. Each benchmark page has the commands.
Why do some pages link to results instead of quoting them?
The head-to-heads with other libraries are published in the open, with live dashboards and result files, so that is where the numbers are read. A page here only quotes a comparison number once it has been checked against the latest published run.
Measured. Now try it on your own data.
Free tier · No credit card to start
or run it yourself: pip install infino · Apache-2.0 · quickstart →