Which format uses the fewest tokens?

Raw CSV is almost always the most token-efficient because it states each column header once and uses single-character delimiters. JSON repeats every key on every row, often costing two to four times more tokens for the same data.

How accurate is the token estimate?

It uses a character-per-token heuristic (~4 characters per token for English-like text) which is close for most tabular data but not exact. For billing-grade numbers, run the chosen representation through your model's real tokenizer.

Why does JSON cost so much more than CSV?

JSON repeats the full key name in quotes for every field in every row, plus braces, colons, and commas. For a 10-column, 100-row table that is 1,000 repeated keys, which dominates the token count versus CSV's single header line.

Is markdown ever worth the extra tokens?

Markdown tables cost more than CSV but can improve a model's accuracy when columns must be reasoned about positionally. For pure data transfer, CSV is cheaper; for human-readable reasoning, the markdown overhead can pay off.

Is my CSV uploaded anywhere?

No. The file is read and tokenized entirely in your browser. Nothing is sent to a server or stored.

CSV / Tabular Data Token Estimator

Send tabular data to an LLM for the fewest tokens

The same table can cost wildly different amounts depending on how you serialise it into a prompt. This estimator takes your CSV and shows the token count for four common representations — raw CSV, JSON, markdown table, and key-value — so you can pick the cheapest one your model handles well.

How the formats compare

Token cost is driven by repeated structural characters. CSV states each column header once and separates values with a single comma. JSON repeats every key on every row:

csv:      name,age\nAda,36
json:     [{"name":"Ada","age":36}]
markdown: | name | age |\n| --- | --- |\n| Ada | 36 |

For wide or tall tables, JSON’s repeated keys can multiply the token count several times over versus CSV for identical data. The estimator approximates tokens at roughly four characters per token, the standard rule of thumb for English-like text.

Tips to cut tabular token cost

Prefer raw CSV for pure data transfer — it is consistently the cheapest.
Send only the columns and rows you need. Dropping unused columns cuts the per-row cost across the whole table.
Abbreviate headers when the model does not need full names; a header that appears once is cheap in CSV but expensive when repeated in JSON.