LLM Token Counter
Count tokens and estimate cost for GPT, Claude, Gemini and Llama prompts, with exact counts where tokenizers are public.
A short shelf for text with a limit on it. Word Counter covers essays, posts and meta descriptions. When the limit is a model's context window, use LLM Token Counter instead, since tokens and words don't line up. The prompt builder and Markdown-to-knowledge-base tool live here because their input is plain writing, not code.
7 tools
Count tokens and estimate cost for GPT, Claude, Gemini and Llama prompts, with exact counts where tokenizers are public.
Structure prompts with a role, task, context, constraints, output format and examples, with reusable variables.
Split Markdown docs into token-sized, heading-aware chunks with metadata, ready for RAG, embeddings or a custom GPT.
Count words and characters as you type, with reading time and length limits.
Change text to UPPER, lower, Title or Sentence case, or camelCase, snake_case and kebab-case.
Compare two texts side by side and see added, removed and changed lines, down to the word.
See how your title and meta description look in Google on desktop and mobile, and where they get cut off.
Other shelves
Beautify or minify HTML, CSS, JavaScript and JSON, and poke at API responses without leaving the tab.
Shrink, convert, crop and resize images before they go on a web page, or find openly licensed ones.
QR codes, page screenshots, meta tags and other one-off jobs that don't fit elsewhere.
For sites and prompts aimed at AI assistants: llms.txt, crawler access, JSON-LD, token counts and MCP config.
Switch a file or value between formats: image types, JSON and XML, hex and RGB.
Merge, split, reorder, sign and convert PDFs in the browser, with no upload for most of them.
Hex/RGB conversion, favicon sets and other small visual assets for a site.
Shrink a video file for email or the web without sending it to a server.
Missing something?
Describe the job and the file types involved. The most requested tools get built first.