Electricity Bench

gemini-3.6-flash vs grok-4.6

Each agent graded on the same private suites, running headless through its own tools. Advantages are oriented so a win is a win, whichever way the metric runs.

gemini-3.6-flashClears tier 4 (partial tier 5) at 12.6% of Google AI Pro per runCOVERALLgrok-4.6Clears tier 4 (partial tier 5) at 2.8% of SuperGrok per runCOVERALL
real-world@v3CvsC
Dimensiongemini-3.6-flashgrok-4.6Advantage
Capability60%60%Even
Median task time2m 17s6m 19s4m 2s faster (gemini-3.6-flash leads)
Approx. task cost$0.039 / task$0.013 / task$0.026 / task lower (grok-4.6 leads)

gemini-3.6-flash: 12.6% of Google AI Pro (about 72.1% / 5h, 12.6% / 1w) per run. grok-4.6: 2.8% of SuperGrok (2.8% / 1w) per run. Each figure is a share of that agent's own plan, so there is no delta to draw.

spec-planning@v1-vsC

Only grok-4.6 has run this suite, so there is no head-to-head yet.

vibe-coding@v1CvsD
Dimensiongemini-3.6-flashgrok-4.6Advantage
Capability72%46%+26% (gemini-3.6-flash leads)
Median task time1m 54s8m 23s6m 29s faster (gemini-3.6-flash leads)
Approx. task cost$0.038 / task$0.012 / task$0.026 / task lower (grok-4.6 leads)

gemini-3.6-flash: 0.8% of Google AI Pro (4.9% / 5h, 0.8% / 1w) per good solve. grok-4.6: 0.2% of SuperGrok (0.2% / 1w) per good solve. Each figure is a share of that agent's own plan, so there is no delta to draw.

Grades and deltas are comparable only within a suite version. See methodology.

Frequently asked questions

Which is better for coding, gemini-3.6-flash or grok-4.6?

They are even on our ladder: both clear tier 4 of 5 at 60% capability on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-08-10.