Grok 4.20 Non-Reasoning vs Llama 3 8B Lunaris
Llama 3 8B Lunaris is cheaper on output tokens, while Grok 4.20 Non-Reasoning offers a larger context window. Choose Grok 4.20 Non-Reasoning or Llama 3 8B Lunaris based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Grok 4.20 Non-Reasoning | Llama 3 8B Lunaris |
|---|---|---|
| Provider | AI/ML API | AI/ML API |
| Input / 1M tokens | $1.63 | $0.07 |
| Output / 1M tokens | $3.25 | $0.07 |
| Context window | 2M | 8K |
| Parameters | — | — |
| Open weights | No | Yes |
| Released | Mar 2026 | Aug 2024 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.