Vendor-neutral · Pricing updated daily

The reference toolkit for LLM engineers.

Token counting, cost estimation and model comparison — computed in your browser, with no affiliation to any model provider. Nothing you paste is uploaded.

Covers OpenAI, Anthropic, Google, Meta, Mistral and Cohere

Tools

Token Counter

Count tokens for OpenAI, Anthropic and Gemini side by side, with the real tokenizers rather than a character estimate.

Open

LLM Cost Estimator

Project monthly API spend across models from your token volume and request count.

Open

Model Comparison

Context windows, max output and per-million pricing for current models, sortable and filterable.

Open

Context Window Visualizer

Soon

See how much of a model’s context window your prompt actually uses.

Prompt Formatter

Soon

Turn a conversation into the exact request shape each provider’s API expects.

JSON Schema Generator

Soon

Generate function-calling schemas for OpenAI, Anthropic and Gemini from one definition.

System Prompt Library

Soon

A curated library of system prompts, searchable by category.

Why the numbers differ

llmtools runs the real tokenizers rather than a chars-divided-by-four rule of thumb, because the two disagree badly on anything that is not plain English. The same paragraph of Chinese can differ by nearly half between OpenAI’s two encodings alone. Where a genuine tokenizer is unavailable — Gemini has no public browser implementation — the figure is marked as an estimate rather than dressed up as a count.