Limits

Limits that apply to Infino Cloud connections. The open-source engine running locally has no request cap, because there is no request: writes go straight to storage.

Request body: 5 MiB

A single data-plane request body is capped at 5 MiB. In practice this applies to append and update, whose bodies carry your rows as Arrow IPC; a request over the cap is rejected with HTTP 413 and nothing is written.

Target 4 MiB or less per batch

The SDK re-encodes rows before sending, so the bytes on the wire are not exactly the bytes you measured client-side. A batch sized right at the cap can pass or fail on that small encoding difference. Keep a margin: aim for at most 4 MiB of Arrow data per append, and treat the last MiB as headroom rather than budget. Rough row budgets, dominated by the vector column (4 bytes per dimension):

row shape	approx. bytes per row	rows per 4 MiB batch
`vector(384)` + ~200 B of text	~1.7 KiB	~2,400
`vector(768)` + ~200 B of text	~3.2 KiB	~1,200
`vector(1536)` + ~200 B of text	~6.2 KiB	~650

Batching and throughput

One append is one atomic commit, so batch size is the main ingest-throughput knob: prefer batches near the 4 MiB target over many small appends, and split large loads into a loop.

BATCH = 2000  # ~4 MiB for 384-dim vectors with short text
for i in range(0, len(rows), BATCH):
    table.append(rows[i : i + BATCH])

const BATCH = 2000; // ~4 MiB for 384-dim vectors with short text
for (let i = 0; i < rows.length; i += BATCH) {
  table.append(rows.slice(i, i + BATCH));
}

Get started

Guides

Infino Cloud

Reference

Request body: 5 MiB

Target 4 MiB or less per batch

Batching and throughput

See also

​Request body: 5 MiB

​Target 4 MiB or less per batch

​Batching and throughput

​See also

Request body: 5 MiB

Target 4 MiB or less per batch

Batching and throughput

See also