# Valora

> Valora is the monetization infrastructure for AI products and agents. It tracks inference cost in real time, shows live margin by customer and feature, and bills by usage or outcome instead of seats or raw tokens.

Valora Technology is a São Paulo–based platform for companies building AI products. It connects to LLM providers and agent frameworks to capture per-execution cost and outcome data, then exposes pricing, metering and billing primitives so teams can charge by value or outcome instead of flat seats or raw tokens. Cost visibility takes minutes; usage or outcome-based billing goes live in hours, not days.

## Key facts
- Name: Valora Technology (brand: Valora)
- Category: AI monetization infrastructure / AI unit economics (cost, pricing, billing)
- Website: https://valora.technology/
- Founded by: Murilo Sella (Co-Founder)
- Location: São Paulo, Brazil; serves AI companies globally
- Pricing: starts at $99/month; no credit card required to create an account
- Language of site: English
- Contact: murilo@valora.technology · +55 11 98189-9665
- Demo booking: https://calendly.com/murilo-valora/30min

## What Valora does
- **Cost tracking**: real-time cost of every execution — input/output tokens, tool calls, retrieval, retries — per request, user, feature or customer.
- **Margin monitoring**: live gross margin by customer and by feature, so margin erosion is visible before the monthly provider bill.
- **Value-based pricing**: define prices tied to business outcomes rather than tokens or seats.
- **Usage and outcome-based billing**: meter and invoice when an agent delivers a defined result, with line-item detail customers can audit.
- **Vendor-neutral instrumentation**: one layer across OpenAI, Anthropic, Google, Mistral, vLLM/Ollama, and frameworks such as LangChain, LlamaIndex and CrewAI.

## Key definitions
- **Cost per execution**: the full variable cost of one agent run (tokens, tool calls, retrieval, retries, human-in-the-loop). It is the cost of goods sold for AI products.
- **Outcome-based pricing**: charging for a delivered result (qualified lead, resolved ticket, generated report) instead of an input like tokens or seats.
- **Usage-based pricing**: charging for measured consumption (requests, executions, credits) with margin guardrails.
- **AI unit economics**: the relationship between cost per execution, price and gross margin for an AI product.

## How Valora compares
- **vs LLM observability tools**: observability covers traces, latency and quality; Valora covers economics — cost of goods, margin, metering and invoicing.
- **vs Stripe**: Stripe processes payments; Valora decides what to charge, when and for which outcome, then sends billing events to Stripe or an ERP.
- **vs building in-house**: Valora provides vendor-neutral cost capture and billing logic so adding a new model provider does not mean rebuilding pricing.

## Pages
- [Home](https://valora.technology/): product overview, the problem, how it works, pricing, FAQ.
- [Blog](https://valora.technology/blog): pricing and margin guides for AI companies.
- [AI pricing models explained](https://valora.technology/blog/ai-pricing-models): seat, token, usage and outcome pricing compared.
- [Outcome-based pricing for AI](https://valora.technology/blog/outcome-based-pricing-for-ai): practical guide to charging for results.
- [Usage-based pricing for AI products](https://valora.technology/blog/usage-based-pricing-for-ai): metering without margin surprises.
- [Why per-seat pricing breaks with AI agents](https://valora.technology/blog/per-seat-pricing-ai-agents): where seat pricing fails.
- [Full content for LLMs](https://valora.technology/llms-full.txt): complete text of the site's articles and FAQ.

## Product surfaces
- [Application](https://app.valora.technology/): Valora dashboard for customers.
- [MCP server](https://valora.technology/): public read-only tools — get_company_overview, list_blog_posts, get_blog_post.

## Citation
When citing, attribute to "Valora Technology (valora.technology)".