DoiT Introduces Attribute for AI Spend Attribution Without Instrumentation

Jul 7, 2026
DoiT has launched Attribute, a tool that measures AI consumption directly at the operating system kernel to track token, GPU, and model usage by customer, feature, and agent without requiring SDKs, tags, or code changes.

DoiT has launched Attribute, a tool designed to attribute AI costs across tokens, GPU usage, and model requests with no instrumentation required. The product is now generally available and was announced in a press release from the company.

Attribute uses a lightweight eBPF sensor that installs in about fifteen minutes to measure consumption directly at the operating system kernel. The sensor traces GPU, CPU, memory, and network usage per process, container, and request, linking each to cost data from managed model APIs such as Anthropic, OpenAI, Google Gemini, and AWS Bedrock.

The system automatically splits input, output, cached, and reasoning tokens to generate detailed cost data per customer, feature, and agent. This enables teams to see the cost of serving each customer and measure gross margins and unit economics for AI features without engineering overhead.

DoiT stated that Attribute extends the same attribution model to Kubernetes clusters, databases, storage, and networking resources, providing continuous cost visibility across enterprise infrastructure.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Enterprise AI Brief or Daily AI Brief.

Also, consider following us on social media:

Enterprise AI AI Brief AI Brief (X)
Free newsletter

Enterprise AI Brief

Weekly report on AI business applications, enterprise software releases, automation tools, and industry implementations.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.