Meta API: models and pricing
Pick up to 3 models · pricing, context and capabilities · simulate your product cost
Data updated on September 30, 2026· 217 models
8 Meta chat models available by API, newest first, with price per million tokens, context and capabilities. Data as of 09/30/2026.
| # | Model | Provider | Input $/1M | Output $/1M | Context | Capabilities | Released |
|---|---|---|---|---|---|---|---|
| 1 | Llama Guard 4 12B | Meta | $0.180 | $0.180 | 164 K | 04/30/2025 | |
| 2 | Llama 4 Maverick | Meta | $0.188 | $0.653 | 1 M | 04/05/2025 | |
| 3 | Llama 4 Scout | Meta | $0.100 | $0.300 | 1.3 M | 01/01/2025 | |
| 4 | Llama 3.3 70B Instruct | Meta | $0.100 | $0.320 | 131 K | 12/06/2024 | |
| 5 | Llama 3.2 1B Instruct | Meta | $0.027 | $0.201 | 60 K | 09/25/2024 | |
| 6 | Llama 3.2 3B Instruct | Meta | $0.050 | $0.330 | 131 K | 09/18/2024 | |
| 7 | Llama 3.1 8B Instruct | Meta | $0.050 | $0.080 | 131 K | 07/23/2024 | |
| 8 | Llama 3.1 70B Instruct | Meta | $0.400 | $0.400 | 131 K | 07/23/2024 |
Frequently asked questions
How much does the Meta API cost?
It depends on the model: from $0.027 to $0.400 per million input tokens. The cheapest is Llama 3.2 1B Instruct ($0.027 / $0.201); the newest is Llama Guard 4 12B ($0.180 / $0.180). Data as of 09/30/2026.
How many Meta models are available by API?
8 chat models in the catalog, as of 09/30/2026. Each one has its own sheet with pricing, context and capabilities.
Where do the prices come from and how often are they updated?
From the public catalogs of OpenRouter, models.dev and LiteLLM, which publish each provider's list prices. They sync once a day and the last update date is shown above the comparison.
What does price per million tokens mean?
It is what the provider charges per million tokens you send (input) or the model generates (output). A token is roughly 4 characters. Output almost always costs more than input.
Sources: OpenRouter, models.dev and LiteLLM. Prices in USD per million tokens, list price, no volume discounts. Prompt tokens are an estimate; each provider's tokenizer may differ.