# Alye

> Alye is an LLM observability tool that records every model call your services make and shows what it cost, how long it took and whether it worked, by service, model and customer, with alerts by email, Slack and Microsoft Teams.

Alye (LLM observability for cost, latency and errors) is made by Onruyl Private Limited. Website: https://www.alye.ai. App: https://app.alye.ai.

Key facts:
- Records every LLM call from Python (3.10+), Node.js (20+) and Java (11+) services with two lines of code; LangChain, LangChain4j, Express and NestJS integrations; accepts OpenTelemetry GenAI spans over OTLP/HTTP JSON.
- No proxy: the SDK wraps the official OpenAI, Anthropic and Google GenAI clients and sends records in the background.
- Providers: OpenAI, Azure OpenAI, Anthropic, Google Gemini and Vertex AI, Amazon Bedrock, DeepSeek, Groq, Mistral, OpenRouter.
- Cost by service, model and customer, priced per region and marketplace (Azure OpenAI, Vertex AI, Bedrock) and with your own prices.
- Alerts every minute on spend, monthly forecast, error rate, p95 latency, call volume and silent services, by email, Slack or Microsoft Teams.
- Prompts are not stored unless turned on; personal data is masked inside your service first.
- Pricing: Free $0 (forever); Team $49 (per month, or $39 paid yearly); Enterprise Custom (from $500 per month). No per-seat pricing.

## Guides and pages

- [What is LLM observability? A practical guide](https://www.alye.ai/llm-observability): Definition, what to measure, SDK vs proxy vs OpenTelemetry, how to choose a tool.
- [How to track LLM costs by service, model and customer](https://www.alye.ai/llm-cost-tracking): How cost per call is calculated, regional prices, per-customer spend, budgets and alerts.
- [OpenAI API cost tracking by feature, model and customer](https://www.alye.ai/openai-cost-tracking): How OpenAI bills tokens and how to see cost per feature and customer.
- [Claude API cost tracking: Anthropic, Bedrock and Vertex AI](https://www.alye.ai/anthropic-cost-tracking): Claude billing including prompt caching, and regional prices on Bedrock and Vertex AI.
- [Azure OpenAI cost tracking across deployment types](https://www.alye.ai/azure-openai-cost-tracking): Global, Data Zone and Regional pricing, and how to track it per call.
- [Alye vs Langfuse](https://www.alye.ai/compare/langfuse): Pricing, what is counted, regional prices, alerts, self-hosting and evaluation compared.
- [Alye vs Helicone](https://www.alye.ai/compare/helicone): SDK vs proxy, plans, regional prices, caching and alerts compared.
- [Alye vs LangSmith](https://www.alye.ai/compare/langsmith): Pricing, per-seat fees, retention, alerts and LangChain fit compared.
- [LLM API pricing](https://www.alye.ai/models): Input, cached input and output prices per million tokens for every current model, by provider, with regional prices and retirement dates.
- [Free LLM price API](https://www.alye.ai/llm-price-api): The same prices as JSON (GET /api/prices, no key, CORS open), an embeddable price table and README badges; free with a link back.
- [Resources](https://www.alye.ai/resources): Every guide, comparison and SDK link in one place.
- [Pricing](https://www.alye.ai/pricing): Free, Team and Enterprise plans.
- [Docs](https://www.alye.ai/docs): Python, Node.js and Java SDKs, LangChain, LangChain4j, Express, NestJS, customer tags, traces, alerts, OpenTelemetry, settings.

## Blog and whitepapers

- [Why your Azure OpenAI and Bedrock bills run 10% above list price](https://www.alye.ai/blog/azure-bedrock-regional-pricing): Data zones, regional endpoints and cross-region profiles cost more than the global price. What the uplift is, where it applies and how to budget for it.
- [Prompt caching: when it saves money and when it costs more](https://www.alye.ai/blog/prompt-caching-math): Cache writes cost more than plain input and reads cost a tenth. The break-even point, worked examples for Claude and OpenAI, and how to tell which way your traffic goes.
- [Introducing Alye: know what every LLM call costs](https://www.alye.ai/blog/introducing-alye): Why we built Alye, what it does, what it deliberately doesn't do, and how to try it on one service in five minutes.
- [The LLM cost control playbook](https://www.alye.ai/whitepapers/llm-cost-control-playbook): A practical guide for engineering and finance teams: how to attribute, price, budget and reduce LLM spend, with a checklist you can run this quarter.

## Optional

- [Full text for AI assistants](https://www.alye.ai/llms-full.txt): every fact and answer on this site in one file.
- [Privacy policy](https://www.alye.ai/privacy)
- [Terms of service](https://www.alye.ai/terms)