Meta Llama models
llm11 can call 8 Meta Llama models through one API, with Jev, TypeSafe AI's System One model, deciding which one answers each request. Blended prices run from $0.057 to $0.400 per million tokens. The newest is Llama Guard 4 12B, released 2025-04-30.
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| Llama Guard 4 12B | $0.180 | $0.180 | 164K | |
| Llama 4 Maverick | $0.188 | $0.652 | 1.05M | |
| Llama 4 Scout | $0.100 | $0.300 | 1.31M | |
| Llama 3.3 70B Instruct | $0.100 | $0.320 | 131K | |
| Llama 3.2 1B Instruct | $0.027 | $0.201 | 60K | |
| Llama 3.2 3B Instruct | $0.050 | $0.330 | 131K | |
| Llama 3.1 70B Instruct | $0.400 | $0.400 | 131K | |
| Llama 3.1 8B Instruct | $0.050 | $0.080 | 131K |
Read from the live catalogue and refreshed daily. Prices are per million tokens.
Questions
- Which Meta models can I call through llm11?
- 8 today, listed with live prices on this page. The list follows the upstream catalogue, so a new Meta model appears here without anyone editing the page.
- What is the cheapest and the most expensive Meta model?
- By blended price, weighting three parts input to one part output: Llama 3.1 8B Instruct is cheapest at $0.057 per million tokens, and Llama 3.1 70B Instruct is the most expensive at $0.400.
- What is the newest Meta model?
- Llama Guard 4 12B, released 2025-04-30. Call it with model set to meta-llama/llama-guard-4-12b.