token-counting

See what Claude Code and Codex actually send to the API — and what each part costs.

AI API gateway that ends manual channel switching with smart routing, auto failover, exponential cooldown, multi-URL scheduling, live request monitoring and soft-error detection.

Local LLM cost-tracking proxy for OpenAI, Anthropic, Gemini, and pinned OpenRouter calls with token usage, failure, and billing-integrity receipts.

Small, independent TypeScript packages for LLM plumbing — token budgets, streaming JSON repair, cost accounting, retries, embedding caches. No provider SDKs.

Code intelligence for agents: find the code that matters and keep your context window and tokens lean.

TokenMap is a desktop app for treemap-based codebase analysis by tokens, size, complexity, hotspots, and refactor priority

🚀 Intelligent Claude Code status line with multi-provider AI support, real-time token counting, and universal model compatibility. Supports Claude (Sonnet 4: 1M, 3.5: 200K), OpenAI (GPT-4.1: 1M, 4o: 128K), Gemini (1.5 Pro: 2M, 2.x: 1M), and xAI Grok (3: 1M, 4: 256K) with verified 2025 context limits.

ZAI LLM reverse proxy and metrics dashboard

A local proxy that converts websites and APIs to clean Markdown. Convert HTML pages, JSON APIs, and dynamic sites. Get token counts for LLM budgeting.

Token Optimization for Context Engineers. 4.8 KB WASM. Sub-millisecond. Zero dependencies.

Lightweight token tracking, cost management, and budget enforcement for LLM API calls

Pure-Go LLM tokenizer and tiktoken-compatible token counter for OpenAI BPE, WordPiece, SentencePiece, Gemini, Llama, Mistral, and Hugging Face adapters.

ttok-style token counting for Amazon Bedrock

A high-performance, multi-agent observability engine designed for the Model Context Protocol (MCP). It provides a non-blocking, transparent proxy layer that implements deterministic token attribution, real-time context-window alerting, and heuristic-driven static analysis to optimize LLM metadata overhead at scale.

A CLI tool to convert your codebase into a single LLM prompt with source tree, prompt templating, and token counting.

Local Docker-first AI traffic proxy and operator console.

A blazing-fast BPE tokenizer for LLMs. Drop-in tiktoken replacement, 20-80x faster.

High accuracy token counting without the vocabulary.

.NET library for accurate token counting, cost calculation, and session-based usage tracking across 12 LLM providers including OpenAI, Anthropic, Google, Azure, and more.