Home / Token cost / Jina AI

The Jina AI MCP server costs 3,840 tokens before the first call.

Every request your agent makes carries every tool definition this server exposes — context your code, documents and conversation can't use, mostly for tools the agent never calls. You don't need them all in the window, and you don't have to pay for them.

QUICK ANSWER The Jina AI MCP server's 21 tool definitions consume 3,840 tokens — 1.9% of a 200k context window, and 2.1× the median MCP server (1,860 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #1129 of 3,105 measured servers · refreshed every build Method →

What that costs before your agent starts working.

Tool definitions are overhead: they occupy context on every request and compete with your code, documents and conversation history for the same window.

200K WINDOW 1.9%
1M WINDOW 0.4%

Corpus context: Jina AI ranks #1129 of 3,105 measured MCP servers by definition cost. The median is 1,860 tokens, p90 is 7,924, and the heaviest (Fusionauth) is 183,337 — 92% of a 200k window on its own. New to this? See MCP token cost and context window in the glossary.

Where the 3,840 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 183 tokens per tool.

ToolCategoryTokens% of server
search_images Read 290 7.6%
search_web Read 288 7.5%
parallel_search_web Read 282 7.3%
search_arxiv Read 226 5.9%
parallel_read_url Read 225 5.9%
search_ssrn Read 223 5.8%
parallel_search_arxiv Read 217 5.7%
parallel_search_ssrn Read 217 5.7%
search_jina_blog Read 211 5.5%
search_bibtex Read 202 5.3%
read_url Read 193 5.0%
extract_pdf Read 187 4.9%
classify_text Read 177 4.6%
capture_screenshot_url Read 162 4.2%
sort_by_relevance Read 157 4.1%
deduplicate_images Read 143 3.7%
deduplicate_strings Read 127 3.3%
guess_datetime_url Read 116 3.0%
expand_query Read 95 2.5%
primer Read 62 1.6%
show_api_key Read 40 1.0%

Your agent uses a handful of these tools. It pays for all 21.

You don't need all 21 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Jina AI: only the tools you grant are exposed to the agent, the rest never load. A smaller window means a sharper agent — less noise when it picks a tool — and every request costs less:

Grant scopeDefinition costReduction
All 21 tools (no gateway) 3,840 tokens
3 granted tools ~549 tokens −86%
5 granted tools ~914 tokens −76%
10 granted tools ~1,829 tokens −52%
  1. Create a free account and register Jina AI — nothing to install.
  2. Grant only the tools you use — ungranted definitions never enter the context window.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
CUT JINA AI TOKEN COST →

Free to start. No card required.

Jina AI token-cost questions.

How many tokens does the Jina AI MCP server use?+

Its 21 tool definitions total 3,840 tokens — 1.9% of a 200k context window — measured with tiktoken o200k_base over the serialised tools/list payload. Exact counts vary slightly by client and model.

Why does Jina AI consume tokens before I send a message?+

MCP clients load every connected server's tool definitions — name, description, and input schema — into the model's context so it knows what it can call. That payload is charged against your context window on every request, whether or not a tool is used.

How do I reduce Jina AI's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Jina AI to only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. A grant of 3 typical tools costs roughly 549 tokens, a 86% reduction.

Does deferred tool loading fix this?+

Partially, in some clients. Claude Code defers MCP tool schemas behind a tool-search step by default, and VS Code has experimental grouping — but you still pay tokens per search and reload, and Cursor, Windsurf and Gemini CLI load definitions upfront. Reducing the exposed tool set cuts the cost in every client.

How these numbers were measured.

01
Serialisation

Each tool is serialised as a tools/list entry — name, description, input schema — from the schemas in the PolicyLayer scan database. Clients differ slightly in framing, so treat counts as close estimates.

02
Tokeniser

tiktoken o200k_base (GPT-4o/o-series). Anthropic's current tokeniser isn't published, so Claude's exact counts will differ; for English text and JSON schemas the totals are close enough to treat these as estimates.

03
Deferred loading

Some clients now defer schema loading (Claude Code's tool search; VS Code experimental grouping). You still pay per search and reload — and Cursor, Windsurf and Gemini CLI load everything upfront.

Computed 12-06-2026 from the PolicyLayer scan database over all 21 catalogued Jina AI tools. Counts refresh with every site build.

Expose only the tools you use — the rest never enter your context.

A PolicyLayer grant scopes Jina AI to the tools you actually allow. Ungranted definitions never load, and every call that does run is checked against policy first.

Free to start. No card required.

43,000+ MCP servers and 220,000+ tools scanned and risk-classified.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.