← Free tools

Free tool · #2

LLM Token Counter

Count the tokens in any prompt — exact for OpenAI models, estimated for Claude and Gemini.

Prices verified 15 Jul 2026 · 131 models

Prompt

701 models

Token count

29tokens

exacttiktoken o200k_base

≈ $0.000073 to send as input · openai · gpt-4o

Context window
128K in · 16.4K out
Tokenizer
openai_tiktoken
Function callingPrompt cachingBatch API
characters
131
words
22
tokens / word
1.32

Token map — how o200k_base splits your text

Input cost at scale — this prompt on gpt-4o

VolumeTokensInput cost
1 call29$0.000073
1K calls29,000$0.07
100K calls2,900,000$7.25
1M calls29,000,000$72.50

This prompt priced across every model — cheapest first

Columns
Coherecommand-r7b-12-2024~33$0.04$0.0000012
Azureministral-3b~33$0.04$0.0000013
DeepInfraQwen2.5-7B-Instruct~33$0.04$0.0000013
OpenAIgpt-5-nano29$0.05$0.0000014
DeepInfraMistral-Small-24B-Instruct-2501~33$0.05$0.0000016
DeepInfraphi-4~33$0.07$0.0000023
Googlegemini-2.0-flash-lite~33$0.08$0.0000025
OpenAIgpt-4.1-nano29$0.10$0.0000029
AWS Bedrockvoxtral-small-24b-2507~33$0.10$0.0000033
Googlegemini-2.0-flash~33$0.10$0.0000033
Googlegemini-2.5-flash-lite~33$0.10$0.0000033
Mistralmistral-small~33$0.10$0.0000033
OpenRouterministral-3b-2512~33$0.10$0.0000033
OpenRoutermistral-small-3.2-24b-instruct~33$0.10$0.0000033
OpenRouterui-tars-1.5-7b~33$0.10$0.0000033
Vercel AI Gatewayllama-4-scout~33$0.10$0.0000033
OpenAIgpt-4o-mini29$0.15$0.0000043
OpenAIgpt-4o-mini-search-preview29$0.15$0.0000043
DeepSeekdeepseek-v4-flash~33$0.14$0.0000046
Azuremistral-nemo~33$0.15$0.0000050
Coherecommand-r-08-2024~33$0.15$0.0000050
Mistralministral-8b-2512~33$0.15$0.0000050
Vercel AI Gatewaypixtral-12b~33$0.15$0.0000050
OpenAIgpt-5.4-nano29$0.20$0.0000058
DeepInfraLlama-Guard-4-12B~33$0.18$0.0000059
1–25 of 701 models1 / 29
Use this tool from code — or hand it to your agentagent prompt · REST · MCP

Hand this tool to your coding agent

Paste into Claude Code, Cursor, Codex, or any agent. It learns to call this tool over REST or MCP, reaches for it when you ask, and keeps our sourced/pending labels honest.

You can use telemeter.ai's free "LLM Token Counter" — a keyless, read-only tool (no signup), available as an HTTP API and an MCP tool. What it does: Count the tokens in any prompt — exact for OpenAI models, estimated for Claude and Gemini.

Whenever I ask you something this tool answers, call it and use its result instead of estimating yourself.

REST:
  curl -sX POST https://telemeter.ai/api/tools/llm-token-counter -H 'content-type: application/json' -d '{"text":"Summarize the following support ticket in one sentence.","model":"gpt-4o"}'

MCP: the telemeter MCP server at https://telemeter.ai/mcp exposes this as the tool "llm-token-counter" — call it with tools/call.

The response is JSON: { status, data, summary, provenance }. "provenance" carries the source and the date the figure was verified. If status is "pending" or "stale", tell me the value isn't currently verified rather than presenting it as fact.

Or call it directly. This tool runs behind one handler with three front doors — this page, a REST endpoint, and an MCP tool. Same result, read-only, no key. The example below is a request that works.

REST
curl -sX POST https://telemeter.ai/api/tools/llm-token-counter \
  -H 'content-type: application/json' \
  -d '{"text":"Summarize the following support ticket in one sentence.","model":"gpt-4o"}'
MCP — tools/call at https://telemeter.ai/mcp
curl -sX POST https://telemeter.ai/mcp \
  -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"llm-token-counter","arguments":{"text":"Summarize the following support ticket in one sentence.","model":"gpt-4o"}}}'
Input schema (JSON Schema)
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "maxLength": 600000
    },
    "model": {
      "type": "string",
      "minLength": 1
    }
  },
  "required": [
    "text",
    "model"
  ],
  "additionalProperties": false
}

Want this across your whole stack?

Telemeter tracks real cost, latency, and quality on every production agent run — on the same verified pricing behind this tool.

No account, no cookies, nothing stored — every answer is computed from a sourced registry and never logged. Data notice →