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. Grok 4.6 leads overall, 86.6 to 83.4.
Grok 4.6 wins this one comfortably. It sits at #26 with 2035.3 Elo against MiniMax M3 at #42 with 1781.0, and the confidence intervals don't come close to touching, so the gap is real. Overall it's 86.61 to 83.44, a 3.17-point spread, and Grok takes most of the metrics I care about: tone and voice 87.86 vs 83.34, substance 87.53 vs 82.23, continuity and emotion 85.78 vs 80.48, and the numbers-and-slop metric by a wide 87.01 to 78.11. MiniMax pushes back in two places: hooks, where it clearly wins 86.51 to Grok's 80.81, and length adherence, 89.72 to 88.03. The records agree with the ranking, 941 wins and 99 losses for Grok against MiniMax's 737 and 294. Then the bill flips the conversation. MiniMax is open weights and costs about two cents per script against roughly twenty cents for Grok, close to 9x cheaper. If the script ships close to as-written, pay for Grok. If it's a volume pass you'll rewrite anyway, the two-cent draft is hard to ignore.
Pick Grok 4.6 if you want the clearly stronger writer on tone, substance, and continuity and can pay roughly twenty cents per script for it.
Pick MiniMax M3 if you want open weights at about two cents a script with stronger hooks, and can accept a 3.17-point overall gap on drafts you plan to edit anyway.
Blue bars: Grok 4.6. Orange bars: MiniMax M3. Same 0–100 scale; the bold bar wins that metric.
| Grok 4.6 | MiniMax M3 | |
|---|---|---|
| Overall / 100 | 86.6 | 83.4 |
| Writing Elo | 2035 | 1781 |
| Run-to-run spread (± overall std) | 2.530 | 2.030 |
| Cost per script (USD) | 0.204 | 0.023 |
| Avg latency (s) | 243.1 | 109.9 |
| Open weights | No | Yes |
Full scorecards: Grok 4.6 · MiniMax M3. How scoring works: methodology.
← All comparisons