Grok 4.7

AI reference score 4.81 ± 1.05 Last 4 weekly runs · 12 images NormalUpdated every Sunday · 2026-10-26Top tiers of one model Compare with previous version

Latest samples

Score trend

low Normal

2026-10-08 · 5.052026-10-09 · 4.642026-10-10 · 5.212026-10-11 · 4.082026-10-12 · 5.552026-10-13 · 5.142026-10-14 · 3.762026-10-15 · 5.532026-10-16 · 5.122026-10-17 · 4.162026-10-18 · 5.702026-10-19 · 5.352026-10-20 · 4.322026-10-21 · 6.402026-10-22 · 5.042026-10-23 · 5.592026-10-24 · 3.792026-10-25 · 4.712026-10-26 · 5.13

19 records in the last 30 days

Earlier samples

low

Versions by score

  1. Grok 4.6
    Grok 4.6 · high
    high 6.79Last 4 weekly runs · 12 images Default tier for this version
  2. Grok 4.5
    Grok 4.5 · high
    high 6.13Last 4 weekly runs · 12 images Default tier for this version
  3. Grok 4.7
    Grok 4.7 · low
    low 4.81Last 4 weekly runs · 12 images

Scores by criterion

Pelican likeness
5.08
Bicycle correctness
5.42
Riding pose
3.83
Composition
6.67
Obvious errors (higher is better)
4.92

Release notes

Original (Chinese)当天在 Grok Build、Grok API(grok-4-7)及第三方路由/云平台可用。官方 API changelog 同日记为上线。

· View release source ↗