token-metering

Verifiable Usage Receipts for AI APIs. Cryptographic proof that your token bill is real. Ed25519 + Fulcio signing, RFC 6962 Merkle proofs, Trillian Tessera log, C2SP witness cosigning, Rekor public anchoring, Rust tokenizer verification. No blockchain.

Streaming metering engine for LLM token billing | Real-time budget enforcement | Built with Flink

A heads-up display for your AI coding agents. Live token metering, spend and plan limits for Claude Code, Codex CLI, Cursor, Devin and 22 more tools.

Check the fidelity of your inference provider

Serverless metering layer that tracks token usage and cost per AI agent in real time. API Gateway → Lambda → DynamoDB + CloudWatch. Know exactly which agent is spending what — at any scale.

Real-Time Cyberpunk LLM Token Billing Waterfall & Cost Audit Engine — Multi-model price calculations, streaming cost analytics & chronicles.

Self-hosted AI Token API server for AI Coworker: sell AI tokens for your AI App through the AICW Store. OpenAI-compatible, prepaid wallets, your keys stay on your server.

OpenAI-compatible streaming LLM reverse proxy middleware with real-time SSE token metering, dynamic multi-model pricing, client-disconnect safety, and billing callbacks.

High-Throughput Fastify SSE Token Metering & Neural Proxy Monitor — Real-time prompt firewall, circuit breaker & LLM latency telemetry.

Inference cost allocation for autonomous AI agent collaborations — Shapley-fair splitting, congestion pricing, token metering. Part of the Agent Trust Stack.

Enterprise gateway proxy metering model token usage across OpenAI, Anthropic, and Gemini with dynamic team budget controls.

Azure API Management as a centralised AI Gateway for Microsoft Foundry and Azure OpenAI models, with per-application cost attribution and chargeback. Production-grade Terraform: backend pools, circuit breakers, Entra ID auth, token governance, FinOps showback.

Java 原生 LLM 网关：按成本/延迟/质量智能路由，Token 精确计量与配额熔断，进程内部署不需 sidecar。

ClinePass subscription gateway for DeepSeek Harness: registers the plan OpenAI-compatible models as a provider route, exposes a local OpenAI-compatible reverse proxy any client can point at, and turns each call token usage into reference-price cost beside the official 5-hour, weekly and monthly quota windows.

Inference cost allocation for autonomous AI agent collaborations — Shapley-fair splitting, congestion pricing, token metering. Part of the Agent Trust Stack.

Self-host open models on your own GPUs: an OpenAI-compatible gateway with metering and Kubernetes-native deployment.

Durable spend governor for AI agents: a reverse proxy that stops a run the moment it trips a budget.

Private AI FinOps + shadow-AI control plane: meter tokens, enforce budgets & data policy, export audit packs for CFO and CISO. Self-hostable TypeScript.