Voxtral TTS Environmental Impact
Open-weight TTS — 9 languages, zero-shot voice cloning
One query, in context
Energy, carbon and water for a single use. ~30 seconds of generated speech
About 1.3× a Google search and 48% of a minute of laptop use.
17% of a Google search.
About 6.0× a drop of water and 6% of a teaspoon of water.
- Where it runsMistral AI (France) / self-hosted
- Grid carbon intensity50 g CO₂/kWh · 90% renewable
- Compared with a Google searchabout the same as a search (0.3 Wh)
- Water~3,333 queries to fill 1 litre
Where Voxtral TTS sits
#95 of 199 by energy per query. Lighter than 55% of all models. Lit dots are the same kind of model; click any dot to open it.
Want something lighter?
Voxtral TTS is already among the lighter well-known options for this kind of work, so there isn’t a clearly lighter alternative to suggest.
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
- 43.3 Wh
- Carbon
- 2.2 g
- Water
- 32 mL
A month of 5 uses a day, 5 days a week. The energy is about 3.6× a phone charge and 39% of a kettle boil.
Work out your own →- Energy
- 86.6 Wh
- Carbon
- 4.3 g
- Water
- 65 mL
A month of 10 uses a day, 5 days a week. The energy is about 7.2× a phone charge and 79% of a kettle boil.
Work out your own →- Energy
- 346.4 Wh
- Carbon
- 17.3 g
- Water
- 260 mL
A month of 40 uses a day, 5 days a week. The energy is about 3.1× a kettle boil and 5% of a UK home's day.
Work out your own →How does Voxtral TTS compare?
Next to similar models. Bars use a log scale, so each step is a big jump.
Detailed breakdown
Energy
Voxtral TTS is an open-weight text-to-speech model supporting 9 languages and zero-shot voice cloning from 2-3 seconds of audio. At ~0.4 Wh per 30 seconds of speech, it matches ElevenLabs v3 on expressiveness while being fully open and self-hostable.
Power source & carbon
Open-weight and self-hostable. Via Mistral API, runs on France's clean nuclear grid (50 g CO2/kWh).
Water use
At ~0.3 mL per 30 seconds of speech — very low. Zero when self-hosted on consumer hardware.
About Voxtral TTS
Voxtral TTS is an open-source audio model from Mistral AI, released in March 26, 2026, that runs well below the category average for energy consumption at 0.40 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 4B parameters, it open-weight tts — 9 languages, zero-shot voice cloning.
Voxtral TTS benefits from running in France, one of the cleaner grid regions in our dataset at 50 gCO₂/kWh with 90% renewable energy. The same model running in a coal-heavy region would produce significantly more carbon per query.
These figures are estimates derived from hardware specifications and API benchmarks — Mistral AI has not published official energy data for Voxtral TTS. Actual consumption may vary significantly depending on batching, quantisation, and infrastructure optimisations that we cannot observe from outside.
Key insights
- Uses less than a third of the average energy for audio models
- Runs on a 90% renewable grid — among the cleanest AI inference locations
- Open-source weights — can be self-hosted on infrastructure you control
What does your Voxtral TTS usage cost the planet?
Use the calculator to estimate your personal footprint based on how often you use Voxtral TTS.
Frequently asked questions
How much energy does Voxtral TTS use per query?
Each Voxtral TTS query consumes approximately 0.40 Wh of energy. This is about the same as a traditional Google search (~0.3 Wh).
What is Voxtral TTS's carbon footprint?
Based on the carbon intensity of Mistral AI (France) / self-hosted, each query produces approximately 0.02 g of CO2. The grid in this region has a carbon intensity of 50 g CO2/kWh with 90% renewable energy.
How much water does Voxtral TTS use?
Each query consumes approximately 0.30 mL of water, primarily used for cooling the data centers that process the request.
How does Voxtral TTS compare to a Google search?
A Voxtral TTS query uses about the same as a Google search in terms of energy. A Google search uses approximately 0.3 Wh, while Voxtral TTS uses 0.40 Wh.
Technical details
- Architecture
- Neural TTS (open-weight)
- Parameters
- 4B
- Release date
- 2026-03-26
- Open source
- Yes