2.5× less energy than GPT-4 Turbo.
GPT-4 Turbo Environmental Impact
OpenAI's optimised GPT-4 variant, still widely deployed
One query, in context
Energy, carbon and water for a single use. Short query (100 input / 300 output tokens)
About 1.7× a Google search and 60% of a minute of laptop use.
About 1.8× a Google search and 10% of 10 m by car.
About 36× a drop of water and 36% of a teaspoon of water.
- Where it runsAzure US East / Sweden
- Grid carbon intensity450 g CO₂/kWh · 25% renewable
- Compared with a Google search2x more than a search (0.3 Wh)
- Water~556 queries to fill 1 litre
Where GPT-4 Turbo sits
#116 of 199 by energy per query. Heavier than 57% of all models. Lit dots are the same kind of model; click any dot to open it.
Want something lighter?
2.1× less energy than GPT-4 Turbo.
7.1× less energy than GPT-4 Turbo.
What a month of use adds up to
Pick how much you use it. Based on one quick use at a time, five days a week.
- Energy
- 54.1 Wh
- Carbon
- 23.8 g
- Water
- 195 mL
A month of 5 chats a day, 5 days a week. The energy is about 4.5× a phone charge and 49% of a kettle boil.
Work out your own →- Energy
- 216.5 Wh
- Carbon
- 95.3 g
- Water
- 779 mL
A month of 20 chats a day, 5 days a week. The energy is about 2.0× a kettle boil and 3% of a UK home's day.
Work out your own →- Energy
- 1.08 kWh
- Carbon
- 476.3 g
- Water
- 3.90 L
A month of 100 chats a day, 5 days a week. The energy is about 9.8× a kettle boil and 15% of a UK home's day.
Work out your own →How does GPT-4 Turbo compare?
Next to similar models. Bars use a log scale, so each step is a big jump.
Detailed breakdown
Energy
GPT-4 Turbo consumes approximately 0.5 Wh per short query — slightly more than GPT-4o, which was optimised as its successor. It remains widely used in production systems due to its established reliability and 128K context window.
Power source & carbon
Runs on Microsoft Azure infrastructure. Same deployment regions as other OpenAI models.
Water use
At ~1.8 mL per query, GPT-4 Turbo has a similar water footprint to GPT-4o.
About GPT-4 Turbo
GPT-4 Turbo is a text and chat model from OpenAI, released in April 9, 2024. OpenAI's optimised GPT-4 variant, still widely deployed. Each query uses 0.50 Wh of energy and produces 0.22 g of CO₂. That's about 1.7x what a Google search consumes.
These figures are estimates derived from hardware specifications and API benchmarks — OpenAI has not published official energy data for GPT-4 Turbo. Actual consumption may vary significantly depending on batching, quantisation, and infrastructure optimisations that we cannot observe from outside.
GPT-4 Turbo in context
The OpenAI GPT family
One chart per tier, on a shared time axis. Hover a dot for details; click to open that version.
Show 3 more tiers
32 versions across 8 tiers. Lightest: GPT-OSS 20B (0.02 Wh); heaviest: GPT Image 2 (5.0 Wh). Charts share a time axis but each tier has its own energy scale, so its trend is clear; the percentage is the change from a tier’s first version to its latest.
What does your GPT-4 Turbo usage cost the planet?
Use the calculator to estimate your personal footprint based on how often you use GPT-4 Turbo.
Frequently asked questions
How much energy does GPT-4 Turbo use per query?
Each GPT-4 Turbo query consumes approximately 0.50 Wh of energy. This is 2x more than a traditional Google search (~0.3 Wh).
What is GPT-4 Turbo's carbon footprint?
Based on the carbon intensity of Azure US East / Sweden, each query produces approximately 0.22 g of CO2. The grid in this region has a carbon intensity of 450 g CO2/kWh with 25% renewable energy.
How much water does GPT-4 Turbo use?
Each query consumes approximately 2 mL of water, primarily used for cooling the data centers that process the request.
How does GPT-4 Turbo compare to a Google search?
A GPT-4 Turbo query uses 2x more than a Google search in terms of energy. A Google search uses approximately 0.3 Wh, while GPT-4 Turbo uses 0.50 Wh.
Technical details
- Architecture
- Multimodal Transformer (decoder-only)
- Context window
- 128,000 tokens
- Release date
- 2024-04-09
- Open source
- No
- Training data cutoff
- 2023-12