Same standard of work. About 5.5× the energy.
Claude Haiku 5.5 scores higher than Claude 3.5 Sonnet on Epoch's Capabilities Index, 158.98 against 130.0, across 58 benchmarks covering coding, agents, long context and writing, while using about 5.5× less energy per answer. Claude Haiku 5.5 has not faced human head-to-head voting, where Claude 3.5 Sonnet carries 87,769 votes, worth weighing.
| Claude Haiku 5.5 | Claude 3.5 Sonnet | |
|---|---|---|
| Energy per answer | 0.515 Wh | 2.86 Wh |
| Carbon released | 0.24 g CO₂ | 1.3 g CO₂ |
| Same carbon as driving | under a metre | 5 metres (17 feet) |
| Energy rank | 98 of 277 | 175 of 277 |
| Capability (58 benchmarks) | 159.0 | 130.0 |
| Human vote rating | - | 1298 |
| Votes cast | - | 87,769 |
| Confidence | Day-one estimate | Inferred |
| Released | 2026-10-07 | 2024-06-20 |
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.