This model has been retired, or is being retired, by its provider (shutdown on 16 August 2026). The sheet stays online for reference and to quantify past usage; the current range is on the model sheets page.
Llama 3.3 70B: cost, energy and CO₂ per token
- Provider
- Meta
- Confidence
- medium
Llama 3.3 70B emits about 241 g of CO₂e, uses 3.8 L of water and 640 Wh of electricity per million generated tokens.
Key figures, per million tokens
public price, in eurosOn the bill
Llama 3.3 70B1M generated tokens€0.69
1M input tokens€0.51
Not on the bill
CO₂e, 1M generated tokensabout 2 km in a combustion car241 g
Water3.8 L
Energy640 Wh
1M input tokens12 g, 184 mL, 31 Wh
Sources and method
Full equations and assumptions in the methodology.
The calculation, per million generated tokens
640 Wh × (1.14 × 0.287 g/Wh + 0.049 g/Wh) = 241 g
- Energy per token
- The per-token energy factor behind the figures above, published with its citation. Jegham et al., arXiv:2505.09598 (v6)
- Infrastructure parameters
- Datacenter parameters applied to this provider (the equations live in the methodology). Jegham et al., arXiv:2505.09598 (v6)
PUE 1.14 Grid carbon intensity (CIF) 0.287 kgCO₂e/kWh On-site water (WUE) 0.12 L/kWh Off-site water (EWIF) 5.11 L/kWh Amortised manufacturing (embodied) 49 gCO₂e/kWh - Price
- Public provider prices. console.groq.com, deprecations
Frequently asked questions
- How much CO₂e for 1 million tokens with Llama 3.3 70B?
- About 241 g per million generated tokens and 12 g per million input tokens (factors tokenclimate-v4-2026-09 of the TokenClimate registry).
- Where do these figures come from?
- Factors published in TokenClimate's versioned registry, each with its citation on this page. The full equations and parameters are in the public methodology, at tokenclimate.com/en/methodology.
- What is the confidence level?
- Medium confidence: derived from a fitted model, or a partial fit completed by a convention.
- How can you measure your organisation's real usage?
- The TokenClimate dashboard connects to your organisation's Anthropic, OpenAI or OpenRouter admin keys (stored encrypted) and measures cost and CO₂e per team and per model. You can see it in the demo: tokenclimate.com/demo. Locally, the open source plugin claude-carbon measures a Claude Code session.
Do you go through OpenRouter?
The dashboard measures everything that goes through your OpenRouter keys, this model included, per team and per model. It also connects directly to Anthropic and OpenAI.