The AI Footprint ← All 266 models

Gemini 3.1 Pro vs GPT-5.3 Codex

Same standard of work. About 7.0× the energy.

Use this
OpenAI
7.11 Wh
≈ 13 metres of driving
Instead of this
Google DeepMind
49.8 Wh
≈ 92 metres of driving

The verdict

GPT-5.3 Codex scores higher than Gemini 3.1 Pro on Epoch's Capabilities Index — 156.58 against 155.0, across 58 benchmarks covering coding, agents, long context and writing — while using about 7.0× less energy per answer. GPT-5.3 Codex has not faced human head-to-head voting, where Gemini 3.1 Pro carries 106,951 votes — worth weighing.

Side by side

GPT-5.3 CodexGemini 3.1 Pro
Energy per answer7.11 Wh49.8 Wh
Same carbon as driving13 metres92 metres
Energy rank195 of 266245 of 266
Capability (58 benchmarks)156.6155.0
Human vote rating1480
Votes cast106,951
ConfidenceDay-one estimateDay-one estimate
Released2026-02-052026-02-19

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 Epoch AI's Capabilities Index, one score combining 58 separate benchmarks; the vote rating comes from blind human comparisons. The full method is here.

Share this

Data built 2026-09-20 · Gemini 3.1 Pro in detail · GPT-5.3 Codex in detail · all 266 models · compare any two models yourself