Text / Chat

Phi-4 Reasoning Environmental Impact

StandardEstimated

14B reasoning model that outperforms o1-mini and DeepSeek-R1-Distill-70B

Architecture
Dense Transformer with chain-of-thought reasoning
Parameters
14B
Context
128,000 tokens
Provider
Microsoft
1.8 Wh
Energy per query
0.79 g
CO₂ per query
7 mL
Water per query
6x more than
vs Google search

Energy per query

1.8 Wh

6x more than a Google search (0.3 Wh)

CO2 per query

0.79 g

Global Average grid (475 gCO₂/kWh)

Water per query

7 mL

~154 queries to fill 1 litre

Processing location

Self-hosted / Azure

Provider

Microsoft

Category

Text / Chat

Grid carbon intensity

475 g CO2/kWh (27% renewable)

How does Phi-4 Reasoning compare?

Ranked #83 of 152 models by energy per query

0 Wh0.45 Wh0.9 Wh1.35 Wh1.8 WhLLaMA 3.2 1BGemini 1.5 ProGPT-4.1 NanoPhi-4 ReasoningGoogle search (0.3 Wh)

Detailed Breakdown

Energy Consumption

Phi-4 Reasoning is a 14B model that outperforms o1-mini and DeepSeek-R1-Distill-70B on reasoning benchmarks. At ~1.8 Wh per reasoning query, the extended thinking adds significant overhead versus standard Phi-4 (~0.35 Wh). The Reasoning Plus variant uses even more compute for harder problems.

Power Source & Carbon

Open-source. Can be self-hosted on a single GPU, making it one of the most accessible reasoning models.

Water Usage

At ~6.5 mL per reasoning query. Standard non-reasoning use consumes much less.

About Phi-4 Reasoning

Phi-4 Reasoning is an open-source text and chat model from Microsoft, released in May 10, 2025, that runs below the category average for energy consumption at 1.8 Wh per query. Because its weights are publicly available, it can be self-hosted on any infrastructure — meaning its carbon footprint depends entirely on where and how you choose to run it. At 14B parameters, it 14b reasoning model that outperforms o1-mini and deepseek-r1-distill-70b.

These figures are estimates derived from hardware specifications and API benchmarks — Microsoft has not published official energy data for Phi-4 Reasoning. Actual consumption may vary significantly depending on batching, quantisation, and infrastructure optimisations that we cannot observe from outside.

Phi-4 Reasoning in Context

16.4 kWh
per year

Your yearly Phi-4 Reasoning footprint

At 25 queries per day, your annual Phi-4 Reasoning usage consumes 16.4 kWh — roughly what a fridge uses in a month. That produces 7.2 kg of CO₂.

Key Insights

Open-source weights — can be self-hosted on infrastructure you control

What does your Phi-4 Reasoning usage cost the planet?

Use our calculator to estimate your personal environmental footprint based on how often you use Phi-4 Reasoning.

Calculate My Compute

Frequently Asked Questions

How much energy does Phi-4 Reasoning use per query?

Each Phi-4 Reasoning query consumes approximately 1.8 Wh of energy. This is 6x more than a traditional Google search (~0.3 Wh).

What is Phi-4 Reasoning's carbon footprint?

Based on the carbon intensity of Self-hosted / Azure, each query produces approximately 0.79 g of CO2. The grid in this region has a carbon intensity of 475 g CO2/kWh with 27% renewable energy.

How much water does Phi-4 Reasoning use?

Each query consumes approximately 7 mL of water, primarily used for cooling the data centers that process the request.

How does Phi-4 Reasoning compare to a Google search?

A Phi-4 Reasoning query uses 6x more than a Google search in terms of energy. A Google search uses approximately 0.3 Wh, while Phi-4 Reasoning uses 1.8 Wh.

Technical Details

Architecture

Dense Transformer with chain-of-thought reasoning

Parameters

14B

Context window

128,000 tokens

Release date

2025-05-10

Open source

Yes

Training data cutoff

2025-04