We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
docs(benchmarks): rewrite with the new benchmark-suite results Per-op throughput + latency at 10k, offset-vs-keyset pagination, 1k->10k scaling, concurrency and memory, from the new benchmarkSuite() harness (median of 3). Refresh the methodology/run instructions and trim the future-work list (most items landed).
docs(benchmarks): refresh numbers and note the ops/s fix landed
docs: add Benchmarks page (per-backend throughput + recommendations)