API pricing

# Meta API pricing

> Meta API prices run from $0.05 per million input tokens (Llama 3.1-8B) to $1.25 (Muse Spark 1.1). Its highest-ranked model, Muse Spark 1.3, costs $1.25 input and $4.25 output per million tokens.
- Canonical page: https://noometry.com/llm-pricing/meta
- Last updated: 2026-10-10
- Title: Meta API Pricing (October 2026): Every Model per 1M Tokens

Meta API prices run from $0.05 per million input tokens (Llama 3.1-8B) to $1.25 (Muse Spark 1.1). Its highest-ranked model, Muse Spark 1.3, costs $1.25 input and $4.25 output per million tokens.

Last verified October 10, 2026

API prices per million tokens
|  |  |  | Cached input |  |  |  |
| --- | --- | --- | --- | --- | --- | --- |
| [Llama 3.1-8B](https://noometry.com/models/llama-3-1-8b) (open weights) | $0.05 | $0.08 | — | **$0.0575** | 128K | 22.8 |
| [Llama 3.2 1B](https://noometry.com/models/llama-3-2-1b) (open weights) | $0.027 | $0.20 | — | **$0.0705** | 60K | 19.9 |
| [Llama 3.2 3B](https://noometry.com/models/llama-3-2-3b) (open weights) | $0.05 | $0.33 | — | **$0.12** | 131K | 28.4 |
| [Muse Spark 1.2 Contributor](https://noometry.com/models/muse-spark-1-2-contributor) | $0.10 | $0.20 | $0.002 | **$0.13** | 1.05M | — |
| [Muse Spark 1.3 Contributor](https://noometry.com/models/muse-spark-1-3-contributor) | $0.10 | $0.20 | $0.002 | **$0.13** | 1.05M | — |
| [Llama 3-8B](https://noometry.com/models/llama-3-8b) (open weights) | $0.14 | $0.14 | — | **$0.14** | 8K | 25.3 |
| [Llama 4 Scout](https://noometry.com/models/llama-4-scout) (open weights) | $0.10 | $0.30 | — | **$0.15** | 128K | 27.4 |
| [Llama-3.3-70B-Instruct](https://noometry.com/models/llama-3-3-70b-instruct) (open weights) | $0.10 | $0.32 | — | **$0.16** | 128K | 30.3 |
| [Llama 4 Maverick](https://noometry.com/models/llama-4-maverick) (open weights) | $0.19 | $0.65 | — | **$0.30** | 128K | 30.6 |
| [Llama 3.1-70B](https://noometry.com/models/llama-3-1-70b) (open weights) | $0.40 | $0.40 | — | **$0.40** | 128K | 29.4 |
| [Muse Glimmer 30B](https://noometry.com/models/muse-glimmer-30b) (open weights) | $0.35 | $1.50 | $0.04 | **$0.64** | 131K | — |
| [Muse Spark 1.1](https://noometry.com/models/muse-spark-1-1) | $1.25 | $4.25 | $0.15 | **$2** | 1.05M | 49.9 |
| [Muse Spark 1.2](https://noometry.com/models/muse-spark-1-2) | $1.25 | $4.25 | $0.15 | **$2** | 1.05M | 49.8 |
| [Muse Spark 1.3](https://noometry.com/models/muse-spark-1-3) | $1.25 | $4.25 | $0.15 | **$2** | 1.05M | 54.4 |

## Prices on other platforms

The same models are often sold through clouds and resellers at different rates.

| Model | Route | Input | Output | Checked |
| --- | --- | --- | --- | --- |
| [Muse Spark 1.3](https://noometry.com/models/muse-spark-1-3) | meta | $1.25 | $4.25 | 2026-09-30 |
|  | openrouter | $1.25 | $4.25 | 2026-09-30 |
| [Muse Spark 1.1](https://noometry.com/models/muse-spark-1-1) | meta | $1.25 | $4.25 | 2026-09-30 |
|  | openrouter | $1.25 | $4.25 | 2026-09-30 |
| [Muse Spark 1.2](https://noometry.com/models/muse-spark-1-2) | meta | $1.25 | $4.25 | 2026-09-30 |
|  | openrouter | $1.25 | $4.25 | 2026-09-30 |
| [Llama 4 Maverick](https://noometry.com/models/llama-4-maverick) | azure | $0.25 | $1 | 2026-09-30 |
|  | bedrock | $0.24 | $0.97 | 2026-09-30 |
|  | deepinfra | $0.20 | $0.80 | 2026-09-30 |
|  | openrouter | $0.19 | $0.65 | 2026-09-30 |
|  | vertex | $0.35 | $1.15 | 2026-09-30 |
| [Llama-3.3-70B-Instruct](https://noometry.com/models/llama-3-3-70b-instruct) | azure | $0.71 | $0.71 | 2026-09-30 |
|  | bedrock | $0.72 | $0.72 | 2026-09-30 |
|  | deepinfra | $0.10 | $0.32 | 2026-09-30 |
|  | groq | $0.59 | $0.79 | 2026-09-30 |
|  | openrouter | $0.10 | $0.32 | 2026-09-30 |
|  | together | $1.04 | $1.04 | 2026-09-30 |
|  | vertex | $0.72 | $0.72 | 2026-09-30 |
| [Llama 3.1-70B](https://noometry.com/models/llama-3-1-70b) | bedrock | $0.72 | $0.72 | 2026-09-30 |
|  | openrouter | $0.40 | $0.40 | 2026-09-30 |
| [Llama 3.2 3B](https://noometry.com/models/llama-3-2-3b) | openrouter | $0.05 | $0.33 | 2026-09-30 |
| [Llama 4 Scout](https://noometry.com/models/llama-4-scout) | azure | $0.20 | $0.78 | 2026-09-30 |
|  | bedrock | $0.17 | $0.66 | 2026-09-30 |
|  | deepinfra | $0.10 | $0.30 | 2026-09-30 |
|  | openrouter | $0.10 | $0.30 | 2026-09-30 |
| [Llama 3-8B](https://noometry.com/models/llama-3-8b) | together | $0.14 | $0.14 | 2026-09-30 |
| [Llama 3.1-8B](https://noometry.com/models/llama-3-1-8b) | bedrock | $0.22 | $0.22 | 2026-09-30 |
|  | groq | $0.05 | $0.08 | 2026-09-30 |
|  | openrouter | $0.05 | $0.08 | 2026-09-30 |
| [Llama 3.2 1B](https://noometry.com/models/llama-3-2-1b) | openrouter | $0.027 | $0.20 | 2026-09-30 |
| [Muse Glimmer 30B](https://noometry.com/models/muse-glimmer-30b) | openrouter | $0.35 | $1.50 | 2026-09-30 |
| [Muse Spark 1.2 Contributor](https://noometry.com/models/muse-spark-1-2-contributor) | meta | $0.10 | $0.20 | 2026-09-30 |
|  | openrouter | $0.10 | $0.20 | 2026-09-30 |
| [Muse Spark 1.3 Contributor](https://noometry.com/models/muse-spark-1-3-contributor) | meta | $0.10 | $0.20 | 2026-09-30 |
|  | openrouter | $0.10 | $0.20 | 2026-09-30 |

## Frequently asked questions

### How much does the Meta API cost?

Meta API prices run from $0.05 per million input tokens (Llama 3.1-8B) to $1.25 (Muse Spark 1.1). Its highest-ranked model, Muse Spark 1.3, costs $1.25 input and $4.25 output per million tokens.

### Does Meta discount cached input?

Yes. Cached input is billed at a lower rate on 6 of the 14 priced models; the cached rate is listed next to each model.