Energy and carbon attribution

The attribution engine does more than tell you who and how much — it also estimates how much energy each AI invocation consumed and how much carbon it emitted. Energy and carbon are attributed across the same attribution graph as cost, so you can see per-team, per-model, and per-cost-center energy and emissions alongside spend.

This page explains how those estimates are produced, where they come from, and how Venturi keeps them honest.

What Venturi estimates

For every intercepted AI request, the engine attaches an energy and carbon block to the invocation:

Estimate	What it is
Estimated energy (kWh)	The GPU energy consumed by the request.
Energy Score rating (1–5 stars)	The model's relative energy efficiency for its task type.
Estimated carbon (gCO2e)	The emissions from that energy, given the deployment region's grid intensity.

These flow into your dashboards, your FOCUS exports, and the optimization recommendations Venturi makes.

How energy is estimated

Venturi estimates GPU energy from three inputs that it already captures for every invocation:

1. Token counts — how much work the request did.
2. Model identity — which model served it.
3. Per-model energy coefficients — drawn from the AI Energy Score leaderboard.

The AI Energy Score is a standardized, independent benchmark maintained by Hugging Face, Salesforce, Cohere, and Carnegie Mellon University. It measures AI model inference energy on NVIDIA H100 GPUs across ten task types — text generation, summarization, extractive QA, text classification, sentence similarity, image classification, object detection, speech recognition, image generation, and image captioning — and assigns each model a 1-to-5 star rating for relative efficiency within its task category. Venturi vendors this catalog and looks up each invocation's model to derive its energy estimate.

The result records how it was derived (the energy derivation method), so the provenance of every energy number is auditable.

How carbon is estimated

Each invocation with a non-null energy estimate carries an estimated carbon emission, computed as:

estimated_carbon_gco2  =  estimated_energy_kwh  ×  grid_carbon_intensity

Venturi resolves the grid carbon intensity in a clear order of precedence, so the most specific value available is always used:

1. An explicit value supplied for the calculation.
2. A region-specific vendored intensity for the deployment region.
3. A configured deployment value (environment variable or Helm value).
4. The fallback default — ≈402 gCO2e/kWh, an approximate US-grid-average operational emission intensity (EPA eGRID basis).

Set your region for accurate carbon

The default is a transparent US-average fallback. Configuring your actual deployment region's grid intensity — or supplying a region-specific value — makes carbon estimates materially more accurate. Region intensity can be overridden via deployment configuration and takes effect without a restart.

The honest-unknown rule: null, never zero

Venturi only estimates energy and carbon for models it can find in the catalog. For any model that is not in the catalog, both estimates are returned as null — never zero.

Null is not zero

A model absent from the energy catalog produces a null energy estimate and a null carbon estimate. Venturi never substitutes zero, because zero would falsely imply no energy was consumed. This honest-unknown discipline is the same principle that governs attribution confidence: Venturi reports what it does not know rather than fabricating a number.

In dashboards and exports, these nulls are shown with explicit honest-uncertainty treatment so an uncatalogued model is visibly unrated, not silently counted as emission-free.

Where you see energy and carbon

Surface	What it shows
Dashboards	Per-team total energy (kWh), per-model efficiency ranking (kWh per 1,000 output tokens), and per-cost-center energy allocation, alongside the cost views.
FOCUS export	Energy and carbon columns — estimated energy (kWh), Energy Score rating, and carbon (gCO2e) — added to the standard FOCUS export. Carbon is non-null for catalogued models and null for uncatalogued ones.
Energy budgets	Per-team and per-cost-center energy caps (kWh per period), evaluated with the same advisory and gate semantics as cost budgets.
Optimization recommendations	When a more efficient model exists for a task, recommendations include the star ratings of the current and proposed models, the energy differential in kWh per 1,000 requests, and the annualized carbon impact of switching.

Energy is not cost

Never read an energy multiplier as a cost multiplier

Energy efficiency varies enormously between models and modes — the same model can vary by a large factor depending on configuration, and reasoning-enabled operation is far more energy-intensive than standard operation. These figures are energy-transparency context only. Energy is not cost. Never restate an energy multiplier as a cost range — cost comes from the billing and pricing path, energy comes from this engine, and the two are reported separately.

Where to go next

For the full formulas, constants, and provenance behind the numbers above:

• Energy & carbon methodology — the T0–T4 tier ladder, the watt-hour primitive and star bands, the AI Energy Score run-config, the reasoning premium, the five-step carbon resolution chain, and the fuel factors.
• Water methodology — the two-term per-query water model.
• Eco-efficiency methodology — capability per environmental cost, distinct from the energy star rating.
• Energy catalog version history — why a star can change without the energy changing.
• Public AI Energy Index — the public, signed dataset and viewer.
• Energy & carbon FAQ — the common questions, answered.

And the surrounding engine docs:

• The attribution engine — how energy attaches to the same records as cost.
• Confidence and evidence — the honest-unknown discipline applied to attribution.
• The HRE pipeline — how each invocation is resolved before energy is attached.