$ loading_
Count tokens for OpenAI-compatible models or encoding names.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "Token Counter" MCP server from askskill: Run: claude mcp add --transport http 'com-apished-token-counter' 'https://apished.com/mcp/v1/tokens'
Developers can check whether prompts and extra text fit within a model's context window before sending a request. This tool quickly estimates token counts by model or encoding name.
Researchers or content creators can count tokens for long text chunks or variants before deciding how to trim or compress content.
It counts tokens using OpenAI-compatible tiktoken encodings, based on either a model name or an encoding name. It is useful for estimating length and context usage before calling an LLM.
You need to provide the text to count, plus a model name or encoding name. Which models/encodings are supported depends on tiktoken and the tool's implementation.
Count prompt tokens and estimate API costs for LLM chats.
Count tokens and estimate costs across many LLMs for planning and budgeting.
Count tokens, estimate costs, optimize prompts, and compare LLM pricing.
Compress and recover local context for more efficient GPT/Codex token usage.
Lets users choose response depth and token usage before answering.
Estimate LLM token spend for coding tasks with ranges, scenarios, and confidence.