grok-4.20-beta-0309-reasoning vs Grok 4.5
Both have similar output pricing, while grok-4.20-beta-0309-reasoning offers a larger context window. Choose grok-4.20-beta-0309-reasoning or Grok 4.5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | grok-4.20-beta-0309-reasoning | Grok 4.5 |
|---|---|---|
| Provider | 302.AI | SpaceXAI |
| Input / 1M tokens | $2.00 | $2.00 |
| Output / 1M tokens | $6.00 | $6.00 |
| Context window | 2M | 500K |
| Parameters | — | — |
| Open weights | No | — |
| Released | Mar 2026 | — |
| LMArena (Chatbot Arena) Elo score | 1451 | 1448 |
| LMArena WebDev Elo score | 1375 | 1553 |
| LMArena Vision Elo score | 1263 | 1288 |
| GPQA Diamond score | — | 93.4 |
| Epoch Capabilities Index score | — | 154 |
FAQ
Is grok-4.20-beta-0309-reasoning or Grok 4.5 cheaper?
grok-4.20-beta-0309-reasoning and Grok 4.5 have comparable output pricing.
Which has the larger context window, grok-4.20-beta-0309-reasoning or Grok 4.5?
grok-4.20-beta-0309-reasoning has the larger context window (2M tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.