Kimi K3 vs Grok 4.7 (high)
Head-to-head on the Towards AI writing benchmark: same 10 real YouTube scripts, five runs each, scored blind by a three-family judge panel. Kimi K3 leads overall, 88.5 to 87.8. Their per-model Elo intervals overlap on this board.
The verdict
This one is closer than the ranks suggest. Kimi K3 sits at #21 with 2217.0 Elo and 88.48 overall; Grok 4.7 (high) is at #26 with 2186.4 Elo and 87.79 overall. Their 95% Elo intervals overlap (2179.1 to 2248.4 against 2130.8 to 2232.5), so do not read much into the exact order. Where Kimi K3 pulls ahead: Length Adherence (90.07 vs 74.12) and Hook Strength (89.24 vs 88.08). Grok 4.7 (high) still wins on Visual Cue Quality (88.85 vs 85.14) and Slop Score (EQ-Bench + ours) (93.30 vs 91.23), so it is not a clean sweep. The trade is price: Kimi K3 costs about $0.260 per article against $0.311 for Grok 4.7 (high), so you pay less for the higher rank. Kimi K3 publishes open weights; Grok 4.7 (high) does not. Default to Kimi K3. Reach for Grok 4.7 (high) when visual cues and slop score matter more to you than the headline rank.
Pick Kimi K3 for the stronger board result, length adherence, and the hook.
Pick Grok 4.7 (high) for visual cues and slop score.
Reading the Elo intervals. Each range estimates uncertainty around one model's Elo. We repeatedly sample from every model's recorded run scores on each task and recalculate the full ranking. Comparing these separate ranges does not test the Elo gap between two models. The tasks stay fixed, so the ranges do not measure how the ranking would change on new tasks.
Metric by metric
Blue bars: Kimi K3. Orange bars: Grok 4.7 (high). Same 0–100 scale; the bold bar wins that metric.
Everything else that differs
| Kimi K3 | Grok 4.7 (high) | |
|---|---|---|
| Overall / 100 | 88.5 | 87.8 |
| Writing Elo | 2217 | 2186 |
| Run-to-run spread (± overall std) | 1.260 | 2.670 |
| Cost per script (USD) | 0.260 | 0.311 |
| Avg latency (s) | 234.3 | 519.0 |
| Open weights | Yes | No |
Full scorecards: Kimi K3 · Grok 4.7 (high). How scoring works: methodology.
← All comparisons