Same standard of work. About 8.4× the energy.
Qwen3.6 27B scores higher than Claude 3.5 Sonnet on Epoch's Capabilities Index — 146.49 against 130.0, across 58 benchmarks covering coding, agents, long context and writing — while using about 8.4× less energy per answer. Qwen3.6 27B has not faced human head-to-head voting, where Claude 3.5 Sonnet carries 87,694 votes — worth weighing.
| Qwen3.6 27B | Claude 3.5 Sonnet | |
|---|---|---|
| Energy per answer | 0.351 Wh | 2.94 Wh |
| Same carbon as driving | under a metre | 5 metres |
| Energy rank | 88 of 266 | 169 of 266 |
| Capability (58 benchmarks) | 146.5 | 130.0 |
| Human vote rating | — | 1298 |
| Votes cast | — | 87,694 |
| Confidence | Calculated | Inferred |
| Released | 2026-04-22 | 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.