MultimodalLowEstimatedReleased 2024-09-25

LLaMA 3.2 90B Vision Environmental Impact

Large open-source multimodal model rivalling GPT-4V

ArchitectureVision-Language Transformer (decoder-only)Parameters90BContext128,000 tokensProviderMeta

The figures below are for one quick question. A longer chat with follow-ups uses about 1.7 Wh, and an hour or two of agent work about 49.0 Wh.

One query, in context

Energy, carbon and water for a single use. Quick single question (one exchange)

Energy0.49Wh per query

About 1.6× a Google search and 59% of a minute of laptop use.

likely 0.24 Wh – 0.98 Wh
Carbon0.20g per query

About 1.7× a Google search and 10% of 10 m by car.

likely 0.10 g – 0.40 g
Water0.91mL per query

About 18× a drop of water and 18% of a teaspoon of water.

likely 0.46 mL – 2 mL
  • Where it runsSelf-hosted (varies, requires multi-GPU)
  • Grid carbon intensity458 g CO₂/kWh · 32% renewable
  • Compared with a Google search2x more than a search (0.3 Wh)
  • Water~1,099 queries to fill 1 litre

Where LLaMA 3.2 90B Vision sits

#113 of 199 by energy per query. Heavier than 56% of all models. Lit dots are the same kind of model; click any dot to open it.

Google searchKettle boilLLaMA 3.2 1B — 0.03 WhGemini 1.5 Pro — 0.45 WhGPT-4.1 Nano — 0.03 WhGPT-4o — 0.43 WhClaude 3.7 Sonnet — 0.21 WhGPT-4.5 — 1.0 Who3 (Reasoning) — 2.3 WhDeepSeek-R1 — 1.2 WhGrok 2 — 0.50 WhDALL-E 3 — 3.0 WhMidjourney — 7.5 WhStable Diffusion XL — 2.9 WhSora — 500 WhGitHub Copilot — 1.5 WhSuno — 75 WhGPT-4o Mini — 0.04 WhDeepSeek-V3 — 0.24 WhGemini 2.0 Flash — 0.08 WhClaude Opus 4 — 0.39 WhClaude Sonnet 4 — 0.21 WhClaude 3.5 Haiku — 0.06 WhMistral Large 2 — 1.8 WhLLaMA 3.1 70B — 0.38 WhPerplexity AI — 0.50 WhImagen 3 — 2.5 WhFlux.1 Dev — 3.8 WhVeo 2 — 350 WhGemini 2.5 Pro — 0.40 WhGemini 2.5 Flash — 0.24 WhGPT-4.1 — 0.37 WhGPT-4.1 Mini — 0.03 WhLLaMA 4 Scout — 0.11 WhLLaMA 4 Maverick — 0.11 WhQwen 3 235B — 0.16 WhGrok 3 — 0.40 WhGemma 3 27B — 0.16 WhPhi-4 — 0.09 WhCommand R+ — 0.56 WhGPT-4o Image Generation — 4.5 WhAdobe Firefly — 3.2 WhIdeogram 2.0 — 3.5 WhRunway Gen-3 Alpha — 200 WhKling — 250 WhPika — 180 WhLuma Dream Machine — 160 WhUdio — 70 Who1 — 2.3 Who1-mini — 0.20 Who3-mini — 0.20 Who4-mini — 0.20 WhGPT-4 Turbo — 0.50 WhGemini 1.5 Flash — 0.10 WhGemini Nano — 0.01 WhClaude 3 Opus — 0.56 WhClaude 3 Haiku — 0.06 WhLLaMA 3.1 8B — 0.06 WhLLaMA 3.1 405B — 2.1 WhLLaMA 3.3 70B — 0.38 WhLLaMA 3.2 11B Vision — 0.08 WhMistral 7B — 0.06 WhMixtral 8x7B — 0.09 WhMixtral 8x22B — 0.22 WhMistral Small 3 — 0.14 WhCodestral — 0.80 WhAmazon Nova Pro — 0.30 WhAmazon Nova Lite — 0.06 WhAmazon Nova Micro — 0.03 WhERNIE Bot 4.0 — 0.30 WhDoubao Pro — 0.30 WhGLM-4 — 0.19 WhKimi — 0.25 WhJamba 1.5 Large — 1.0 WhStable Diffusion 3.5 — 3.2 WhLeonardo AI — 3.5 WhHaiLuo AI — 220 WhStable Video Diffusion — 150 WhCursor — 1.2 WhAmazon Q Developer — 1.0 WhClaude Code — 2.0 WhWindsurf (Codeium) — 0.80 WhReplit Agent — 1.5 WhWhisper — 0.50 WhOpenAI TTS — 0.30 WhElevenLabs — 1.5 WhNotebookLM Audio Overview — 15 WhClaude Sonnet 4.5 — 0.34 WhClaude Opus 4.5 — 0.42 WhClaude Opus 4.6 — 0.40 WhClaude Sonnet 4.6 — 0.20 WhGPT-5 — 0.44 WhGPT-5 Mini — 0.05 WhGPT-5.2 — 0.61 WhGPT-5.2 Codex — 0.61 WhGPT-5.4 — 0.60 WhGPT-5.4 Mini — 0.03 Who3-pro — 6.0 WhGPT Image 1.5 — 3.8 WhSora 2 — 600 WhGemini 3 Pro — 0.40 WhGemini 3 Flash — 0.15 WhGemini 3.1 Pro — 0.42 WhGemma 3 4B — 0.04 WhGemma 3 12B — 0.08 WhImagen 4 — 2.8 WhVeo 3 — 400 WhVeo 3.1 — 2.4 kWhGrok 4 — 0.60 WhGrok 4.1 — 0.50 WhDeepSeek-R1-0528 — 1.2 WhDeepSeek-V3.1 — 0.24 WhQwen3.5 (397B-A17B) — 0.13 WhQwen3.5 35B-A3B — 0.04 WhMistral Large 3 — 1.5 WhDevstral 2 — 0.60 WhVoxtral TTS — 0.40 WhKimi K2.5 — 0.30 WhGLM-5 — 0.22 WhDoubao 2.0 — 0.30 WhCommand A — 0.60 WhPhi-4 Mini — 0.04 WhPhi-4 Reasoning — 0.27 WhAmazon Nova 2 Pro — 0.35 WhMidjourney V7 — 8.0 WhFlux 2 Pro — 5.0 WhRunway Gen-4.5 — 250 WhKling 3.0 — 280 WhDevin — 3.0 WhElevenLabs v3 — 1.8 WhSuno v5.5 — 80 WhClaude Haiku 4.5 — 0.02 WhGPT-5.3 Codex — 0.45 WhGPT-5.4 Nano — 0.03 WhGPT-OSS 120B — 0.03 WhGPT-OSS 20B — 0.02 WhGemini 3.1 Flash Lite — 0.06 WhLyria 3 Pro — 50 WhDeepSeek-V3.2 — 0.25 WhGrok 4.20 — 0.50 WhQwen3-Coder 480B — 0.20 WhQwen 3.5 Omni — 0.20 WhNemotron 3 Super — 0.45 WhDoubao Seed 2.0 Pro — 0.35 WhSeedance 2.0 — 280 WhGLM-5.1 — 0.22 WhERNIE 5.0 — 0.35 WhMistral Small 4 — 0.40 WhPhi-4 Multimodal — 0.05 WhMidjourney V8 Alpha — 5.0 WhRecraft V4 — 3.0 WhLuma Ray 3 — 200 WhPika 2.5 — 210 WhGPT-5.5 — 0.51 WhGPT-5.5 Instant — 0.06 WhGPT-5.3 Instant — 0.05 WhGPT Image 2 — 5.0 WhClaude Opus 4.7 — 0.44 WhClaude Opus 4.8 — 0.53 WhClaude Fable 5 — 0.79 WhClaude Sonnet 5 — 0.24 WhDeepSeek-V4-Pro — 0.24 WhDeepSeek-V4-Flash — 0.10 WhNano Banana 2 — 1.5 WhQwen3.6 27B — 0.16 WhQwen3.7-Max — 0.45 WhSora 2 Pro — 2.1 kWhClaude Opus 5 — 0.46 WhClaude Fable 5.1 — 0.65 WhClaude Opus 5.5 — 0.45 WhClaude Sonnet 5.5 — 0.24 WhClaude Haiku 5.5 — 0.02 WhClaude Fable 5.2 — 0.60 WhGPT-5.6 Sol — 0.39 WhGPT-5.6 Terra — 0.16 WhGPT-5.6 Luna — 0.05 WhGPT-6 Astra — 0.90 WhGPT-6 Sol — 0.42 WhGPT-6 Luna — 0.06 WhGPT-6 Instant — 0.07 WhGPT-6.1 Sol — 0.42 WhGPT-6.1 Terra — 0.17 WhGemini 3.5 Flash — 0.24 WhGemini 3.5 Flash-Lite — 0.06 WhGemini 3.6 Flash — 0.21 WhGemini 3.7 Flash — 0.14 WhGemini 3.8 Flash — 0.15 WhGemini 4 Argon — 0.45 WhNano Banana 2.1 — 1.5 WhGrok 4.5 — 0.35 WhGrok 4.7 — 0.40 WhMuse Spark — 0.40 WhMuse Spark Open — 0.22 WhKimi K3 — 0.65 WhGLM-5.3 — 0.22 WhQwen3.8-Max — 0.55 WhDeepSeek-V4.1-Flash — 0.10 WhMistral Large 4 — 1.4 WhKling 4.0 — 300 WhSeedance 2.5 — 290 Wh0.1 Wh1 Wh10 Wh100 Wh1 kWhLLaMA 3.2 90B Vision

Want something lighter?

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
53.0 Wh
Carbon
21.7 g
Water
99 mL

A month of 5 uses a day, 5 days a week. The energy is about 4.4× a phone charge and 48% of a kettle boil.

Work out your own →
Energy
106.1 Wh
Carbon
43.3 g
Water
197 mL

A month of 10 uses a day, 5 days a week. The energy is about 8.8× a phone charge and 96% of a kettle boil.

Work out your own →
Energy
424.3 Wh
Carbon
173.2 g
Water
788 mL

A month of 40 uses a day, 5 days a week. The energy is about 3.9× a kettle boil and 6% of a UK home's day.

Work out your own →

How does LLaMA 3.2 90B Vision compare?

Next to similar models. Bars use a log scale, so each step is a big jump.

LLaMA 3.2 90B Vision0.49 Wh

Detailed breakdown

Energy

We estimate LLaMA 3.2 90B Vision at ~0.49 Wh for a quick single question. The open-weight models CLEER has benchmarked directly follow a clear pattern: roughly 0.02 Wh plus 0.005 Wh per billion parameters active on each token (our fit across 11 models from 3B to 49B active). LLaMA 3.2 90B Vision activates ~90B (it is a dense model, so every parameter is active), which gives ~0.49 Wh when served from a batched data-centre deployment. Image inputs add encoding work on top of this text-only figure. Run locally, the figure depends entirely on your hardware. This is our extrapolation, not a CLEER figure or a measurement. It replaces our earlier ~1.5 Wh estimate, which was built on 2025 benchmark figures for older hardware.

Power source & carbon

Requires cloud GPU clusters for inference due to its size. Available through major cloud providers and specialised inference platforms.

Water use

At ~0.91 mL per quick question, derived from the energy estimate at the ~1.9 L/kWh water factor we apply to this model's infrastructure. Water scales directly with energy, so longer chats, reasoning and agentic sessions use proportionally more.

About LLaMA 3.2 90B Vision

LLaMA 3.2 90B Vision is a multimodal model from Meta, released in September 25, 2024. Large open-source multimodal model rivalling GPT-4V. Each query uses 0.49 Wh of energy and produces 0.20 g of CO₂. That's about 1.6x what a Google search consumes.

These figures are estimates derived from hardware specifications and API benchmarks — Meta has not published official energy data for LLaMA 3.2 90B Vision. Actual consumption may vary significantly depending on batching, quantisation, and infrastructure optimisations that we cannot observe from outside.

LLaMA 3.2 90B Vision in context

90%potential savings

The efficiency alternative

Phi-4 Multimodal performs the same type of task using just 0.05 Wh per query — 90% less energy than LLaMA 3.2 90B Vision. For a user sending 25 queries per day, switching would save 4.0 kWh per year.

Key insights

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

The Meta LLaMA family

One chart per tier, on a shared time axis. Hover a dot for details; click to open that version.

Vision2 models
Google searchLLaMA 3.2 90B Vision: 0.49 Wh · Sept 20240.49 WhLLaMA 3.2 11B Vision: 0.08 Wh · Sept 202420242025
LLaMA 3.2 90B Vision, LLaMA 3.2 11B Vision
Flagship5 models
Google searchLLaMA 3.1 70B: 0.38 Wh · Jul 2024LLaMA 3.1 8B: 0.06 Wh · Jul 2024LLaMA 3.1 405B: 2.1 Wh · Jul 2024LLaMA 3.2 1B: 0.03 Wh · Sept 2024LLaMA 3.3 70B: 0.38 Wh · Dec 202420242025
LLaMA 3.1 70B, LLaMA 3.1 8B, LLaMA 3.1 405B, LLaMA 3.2 1B, LLaMA 3.3 70B
One-off releases2 releases
LLaMA 4 Scout: 0.11 Wh · Apr 2025LLaMA 4 Maverick: 0.11 Wh · Apr 202520242025
LLaMA 4 Scout, LLaMA 4 Maverick

9 versions across 2 tiers. Lightest: LLaMA 3.2 1B (0.03 Wh); heaviest: LLaMA 3.1 405B (2.1 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 LLaMA 3.2 90B Vision usage cost the planet?

Use the calculator to estimate your personal footprint based on how often you use LLaMA 3.2 90B Vision.

Calculate my compute

Frequently asked questions

How much energy does LLaMA 3.2 90B Vision use per query?

Each LLaMA 3.2 90B Vision query consumes approximately 0.49 Wh of energy. This is 2x more than a traditional Google search (~0.3 Wh).

What is LLaMA 3.2 90B Vision's carbon footprint?

Based on the carbon intensity of Self-hosted (varies, requires multi-GPU), each query produces approximately 0.20 g of CO2. The grid in this region has a carbon intensity of 458 g CO2/kWh with 32% renewable energy.

How much water does LLaMA 3.2 90B Vision use?

Each query consumes approximately 0.91 mL of water, primarily used for cooling the data centers that process the request.

How does LLaMA 3.2 90B Vision compare to a Google search?

A LLaMA 3.2 90B Vision query uses 2x more than a Google search in terms of energy. A Google search uses approximately 0.3 Wh, while LLaMA 3.2 90B Vision uses 0.49 Wh.

Technical details

Architecture
Vision-Language Transformer (decoder-only)
Parameters
90B
Context window
128,000 tokens
Release date
2024-09-25
Open source
Yes
Training data cutoff
2024-08

Sources