Same standard of work. About 8.8× the energy.
Claude Sonnet 5.5 scores higher than Gemini 3.1 Pro on Epoch's Capabilities Index, 164.45 against 154.67, across 58 benchmarks covering coding, agents, long context and writing, while using about 8.8× less energy per answer. Claude Sonnet 5.5 has not faced human head-to-head voting, where Gemini 3.1 Pro carries 127,649 votes, worth weighing.
| Claude Sonnet 5.5 | Gemini 3.1 Pro | |
|---|---|---|
| Energy per answer | 5.64 Wh | 49.7 Wh |
| Carbon released | 2.6 g CO₂ | 23 g CO₂ |
| Same carbon as driving | 10 metres (34 feet) | 92 metres (300 feet) |
| Energy rank | 193 of 277 | 260 of 277 |
| Capability (58 benchmarks) | 164.4 | 154.7 |
| Human vote rating | - | 1480 |
| Votes cast | - | 127,649 |
| Confidence | Day-one estimate | Day-one estimate |
| Released | 2026-09-28 | 2026-02-19 |
Energy is calculated the same way for every model on this site, from public data and calibrated against laboratory measurement, accurate to roughly a factor of two to three, which is why we only ever state a difference of 5× or more. Capability is Epoch AI's Capabilities Index, one score combining 58 separate benchmarks; the vote rating comes from blind human comparisons. The full method is here.