Grok 4.20 (Non-Reasoning) vs ALLaM-2-7b
ALLaM-2-7b is cheaper on output tokens, while Grok 4.20 (Non-Reasoning) offers a larger context window. Choose Grok 4.20 (Non-Reasoning) or ALLaM-2-7b based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Grok 4.20 (Non-Reasoning) | ALLaM-2-7b |
|---|---|---|
| Provider | Tempr Gateway | Tempr Gateway |
| Input / 1M tokens | $1.25 | $0.00 |
| Output / 1M tokens | $2.50 | $0.00 |
| Context window | 1M | 4K |
| Parameters | — | — |
| Open weights | No | Yes |
| Released | Mar 2026 | Jan 2025 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.