Kimi K3 vs Grok 4.20 Reasoning
Grok 4.20 Reasoning is cheaper on output tokens, while Grok 4.20 Reasoning offers a larger context window. Choose Kimi K3 or Grok 4.20 Reasoning based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Kimi K3 | Grok 4.20 Reasoning |
|---|---|---|
| Provider | Vercel AI Gateway | Vercel AI Gateway |
| Input / 1M tokens | $3.00 | $1.25 |
| Output / 1M tokens | $15.00 | $2.50 |
| Context window | 1M | 2M |
| Parameters | — | — |
| Open weights | Yes | No |
| Released | Jul 2026 | Mar 2026 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.