Grok 4.20 (Reasoning) vs Claude Opus 4.8
Grok 4.20 (Reasoning) is cheaper on output tokens, while Grok 4.20 (Reasoning) offers a larger context window. Choose Grok 4.20 (Reasoning) or Claude Opus 4.8 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Grok 4.20 (Reasoning) | Claude Opus 4.8 |
|---|---|---|
| Provider | Vertex | Vertex |
| Input / 1M tokens | $1.25 | $5.00 |
| Output / 1M tokens | $2.50 | $25.00 |
| Context window | 2M | 1M |
| Parameters | — | — |
| Open weights | No | No |
| Released | Mar 2026 | May 2026 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.