The AI Footprint ← All 134 models

Claude Sonnet 4.5 vs Qwen3.6 27B

Same standard of work. About 23× the energy.

Use this
Alibaba
0.351 Wh
under 1 m of driving
Instead of this
Anthropic
7.97 Wh
≈ 15 m of driving

The verdict

Qwen3.6 27B scores higher than Claude Sonnet 4.5 on Epoch's Capabilities Index — 147.47 against 147.02, across 58 benchmarks covering coding, agents, long context and writing — while using about 23× less energy per answer. Qwen3.6 27B has not faced human head-to-head voting, where Claude Sonnet 4.5 carries 79,687 votes — worth weighing.

Side by side

Qwen3.6 27BClaude Sonnet 4.5
Energy per answer0.351 Wh7.97 Wh
Same carbon as drivingunder 1 m15 m
Energy rank38 of 13492 of 134
Capability (science questions)84.873.7
Epoch Capabilities Index147.47147.02
Human vote rating1438
Votes cast79,687
ConfidenceCalculatedDay-one estimate
Released2026-04-222025-09-29

What these figures are

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 a science-question test; the Capabilities Index is a composite over 58 benchmarks; the vote rating comes from blind human comparisons. The full method is here.

Share this

Data built 2026-09-02 · Claude Sonnet 4.5 in detail · Qwen3.6 27B in detail · all 134 models · compare any two models yourself