How accurate are these token estimates?

The tool uses a deterministic character-density approach with a fixed confidence level of 0.85, providing a reliable approximation for various LLM architectures.

Which LLM models are supported?

The estimator includes specific modifiers for GPT, Claude, and Llama models to account for their unique tokenization efficiencies.

Can I check if a language is supported before estimating?

Yes, you can use the `validate_language_support` tool to verify if a specific language is included in the supported list.

Char-to-Token Estimator MCP Connector for Claude

A+

Predict token usage for different LLMs using language-specific character ratios.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner

More Details Connect to Claude

This MCP server provides deterministic token estimation by analyzing text density across various languages. It uses specific character-to-token ratios for languages like English, Chinese, and Arabic, and applies model-specific modifiers for GPT, Claude, and Llama architectures. Use estimate_tokens to get a full breakdown of character counts, word counts, and estimated tokens, or get_text_statistics for basic linguistic metrics.

tokensllmtextestimationlinguistics

Connect to Claude

Subscribe on Vinkius, then add this connector to Claude.ai or Claude Code.

① Claude.ai (web app)

Go to Settings → Connectors → Add custom connector
Paste the MCP endpoint URL below

https://edge.vinkius.com/vk_preview_Y2mf9Fr9SfgCCOufW13ktoXXXCGBDxuCrGnCkSZq/mcp

② Claude Code (terminal)

claude mcp add --transport http char-to-token-estimator https://edge.vinkius.com/vk_preview_Y2mf9Fr9SfgCCOufW13ktoXXXCGBDxuCrGnCkSZq/mcp

③ Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "char-to-token-estimator": {
      "url": "https://edge.vinkius.com/vk_preview_Y2mf9Fr9SfgCCOufW13ktoXXXCGBDxuCrGnCkSZq/mcp"
    }
  }
}

Get full access on Vinkius

The preview token above works for testing. Powered by Vinkius.