Gemini 2.5 Flash
Efficient default model; basis of Google's published footprint figure.
Specs
- Provider
- Family
- Gemini
- Context window
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Input price
- $0.3 / 1M
- Output price
- $2.5 / 1M
- Modalities
- text, image, audio, video
- License
- proprietary
- Released
- Apr 17, 2025
- Training cutoff
- Jan 1, 2025
Estimated footprint
Per single text prompt (token-scaled rows assume ~1,000 tokens).
Footprint numbers are rough public estimates that vary by 10–100× across sources and depend on data-centre efficiency and local grid mix. They are shown as a low–high band with a source and date. Don't treat them as measured fact.
Energy
≈ 0.24 Wh
range 0.1–0.5 Wh
~2 s of a 9 W LED bulb
Google, "Measuring the environmental impact of AI inference" (2025) · medium confidence
Water
≈ 0.26 mL
range 0.1–0.6 mL
~0.0 s of a shower
Google, "Measuring the environmental impact of AI inference" (2025) · medium confidence
CO₂e
≈ 0.03 g
range 0.01–0.08 g
~0 m driven in a petrol car
Google, "Measuring the environmental impact of AI inference" (2025) · medium confidence
Metric sources
| Metric | Low–Mid–High | Basis | Confidence | As of | Source |
|---|---|---|---|---|---|
| co2 | 0.01 · 0.03 · 0.08 g | per prompt | medium | Aug 21, 2025 | Google, "Measuring the environmental impact of AI inference" (2025) |
| energy | 0.1 · 0.24 · 0.5 Wh | per prompt | medium | Aug 21, 2025 | Google, "Measuring the environmental impact of AI inference" (2025) |
| water | 0.1 · 0.26 · 0.6 mL | per prompt | medium | Aug 21, 2025 | Google, "Measuring the environmental impact of AI inference" (2025) |
Changelog
- Oct 6, 2026Seeded into PromptLedger reference library.