Free tools for working with AI models
Count tokens, compare model prices, plan context and clean up what AI writes. Every tool runs in your browser, with no sign-up and nothing uploaded.
Prices for 65 models from 6 providers, updated 24 October 2026. Open the full token counter
All tools
Measure
Know how big a prompt is before you send it.
Budget
Work out what a model will cost you, per call and per month.
Clean and check
Fix the output AI gives you before you use it.
Write prompts
Build and refine the instructions you give a model.
Lowest-cost models this week
For a typical chat request of 2,000 input and 400 output tokens. Prices in USD per 1 million tokens.
| Model | Provider | Input | Output | Context |
|---|---|---|---|---|
| GPT-5 Nano | OpenAI | $0.05 | $0.40 | 272K |
| GPT-4.1 Nano | OpenAI | $0.10 | $0.40 | 1M |
| Gemini 2.5 Flash Lite | $0.10 | $0.40 | 1M | |
| GPT-6 Luna | OpenAI | $0.10 | $0.50 | 922K |
| Claude Haiku 5.5 | Anthropic | $0.10 | $0.50 | 1M |
| GPT-4o Mini | OpenAI | $0.15 | $0.60 | 128K |
How this site works
Your text stays on your device. Token counting, JSON repair, text cleaning and every other tool run as code inside your browser. Nothing you paste is sent to us or to an AI provider, so you can use the tools with client emails, contracts and internal documents.
Prices are checked every week. An automated job reads a public, structured model price list, updates the site and records each change on the data updates page. Unusually large changes are held for a person to check before they appear.
Estimates are labelled as estimates. GPT token counts use OpenAI's published tokenizer and are exact. Where a number is approximate, such as Claude or Gemini token counts, the tool says so and explains why.