Llama 4 Scout vs Llama 3.3 70B Instruct
Llama 4 Scout is the cheaper option on input; Llama 4 Scout is cheaper on output by 6%. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Llama 4 Scout | Llama 3.3 70B Instruct | |
|---|---|---|
| Input per 1M tokens | $0.10 | $0.10 |
| Output per 1M tokens | $0.30 | $0.32 |
| Cached input per 1M | — | — |
| Context window | 1.31072M tokens | 131K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | Meta | Meta |
| Released | — | — |
What the numbers say
- Input price gap: 0% ($0.10 vs $0.10).
- Blended three-to-one input/output rate: $0.60 for Llama 4 Scout, $0.62 for Llama 3.3 70B Instruct.
- Context: 1.31072M tokens against 131K tokens.
- Licence: Llama 4 Scout is open weights; Llama 3.3 70B Instruct is open weights.
Sources
Llama 4 Scout specs · Llama 3.3 70B Instruct specs · Full price index