Llama 3.1 70B (Instruct) is cheaper than o3 Mini at $0.59/M vs $1.1/M input tokens.
| Spec | Llama 3.1 70B (Instruct) | o3 Mini |
|---|---|---|
| Input price | $0.59/M tokens | $1.1/M tokens |
| Output price | $0.79/M tokens | $4.4/M tokens |
| Context window | 128K tokens | 64K tokens |
| Parameters | 70B | Not disclosed |
| Weights | Open | Closed |
| Free API (OpenRouter) | No | No |
| Released | Jul 2024 | Dec 2024 |
| MMLU | 86.0% | 85.9% |
Not enough votes to call it. On the specs, Llama 3.1 70B (Instruct) has the edge: bigger model tier, bigger context window.
Llama 3.1 70B (Instruct) costs 5.6x less per token.
Llama 3.1 70B (Instruct) is cheaper on both: 1.9× input, 5.6× output.
Per million tokens. Prices and uptime via OpenRouter, checked 9 Sep 2026.
o3 Mini uses 33.0x more headings