Meta · API pricing
Meta API pricing
3 Meta models on one OpenAI-compatible API, billed per token from a prepaid balance — no subscription. From €0.1512 per 100k input tokens and €0.1512 per 100k output tokens, margin and VAT included. Llama 3.3 70B runs on EU-sovereign infrastructure; Muse Spark 1.1 and Muse Spark 1.2 are fast access, through US providers.
- Models live
- 3
- Input from
- €0.1512
- per 100k tokens
- Output from
- €0.1512
- per 100k tokens
- Largest context
- 1M
- Muse Spark 1.1
Meta API pricing per model
per 100k tokens · margin and VAT included · from our live catalog when this page is built
| Model | Input | Output | Cached in | Context | Max output | Jurisdiction |
|---|---|---|---|---|---|---|
Llama 3.3 70Bllama-3.3-70btools | €0.1512 | €0.1512 | — | — | — | EU sovereign |
Muse Spark 1.1muse-spark-1.1toolsvision | €0.2050 | €0.6956 | €0.0252 | 1M | — | Fast access (US) |
Muse Spark 1.2muse-spark-1.2toolsvision | €0.2050 | €0.6956 | €0.0252 | 1M | — | Fast access (US) |
Input, output and cached input are the prices debited from your prepaid balance per 100,000 tokens, margin and VAT included. Context is the window as served on Frontière AI; max output is the provider's cap on max_tokens (— when it sets none we know of). Tool calling is measured with a real call per model.
Meta API cost calculator
Pick a model and a workload, or type your own token counts: the estimate uses the prices in the table above, margin and VAT included.
Example workload
Roughly 1,000 tokens ≈ 750 English words.
No discounted cache rate on this model: every input token is billed at the input rate.
Per request
€0.0028
Per day
€2.80
Per month
€83.92
30 days
A €10 top-up covers about 3,575 requests · Llama 3.3 70B
Where the money goes
- Input€68.04 · 81%
- Output€15.88 · 19%
Estimate, margin and VAT included, from the prices on this page. You are billed on the tokens the provider reports for each call, debited from a prepaid balance — no subscription.
Same workload on every model
| Model | Per request | Per month | |
|---|---|---|---|
| Llama 3.3 70BEU sovereign | €0.0028 | €83.92 | Selected |
| Muse Spark 1.1Fast access (US) | €0.0055 | €165.26 | |
| Muse Spark 1.2Fast access (US) | €0.0055 | €165.26 |
Call Meta models
The API is OpenAI-compatible: keep your SDK, change base_url and the key, and pass the model id. One key covers the whole catalog.
curl https://getfrontiereai.eu/api/v1/chat/completions \
-H "Authorization: Bearer $FRONTIERE_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "llama-3.3-70b",
"messages": [{"role": "user", "content": "Bonjour"}]
}'import os
from openai import OpenAI
client = OpenAI(
base_url="https://getfrontiereai.eu/api/v1",
api_key=os.environ["FRONTIERE_KEY"],
)
reply = client.chat.completions.create(
model="llama-3.3-70b",
messages=[{"role": "user", "content": "Bonjour"}],
)
print(reply.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://getfrontiereai.eu/api/v1",
apiKey: process.env.FRONTIERE_KEY,
});
const reply = await client.chat.completions.create({
model: "llama-3.3-70b",
messages: [{ role: "user", content: "Bonjour" }],
});
console.log(reply.choices[0].message.content);Frequently asked
How much does the Meta API cost?
On Frontière AI, Meta models cost from €0.1512 to €0.205 per 100,000 input tokens and from €0.1512 to €0.6956 per 100,000 output tokens, depending on the model — margin and VAT included, that is the amount debited from your prepaid balance. The table lists each model's price; the calculator turns it into a monthly cost for your workload.
Which Meta model is the cheapest?
Llama 3.3 70B, the cheapest on both input (€0.1512) and output (€0.1512) per 100k tokens.
Which Meta model has the largest context window?
Muse Spark 1.1, with 1M tokens of context as served on Frontière AI.
Is the Meta API hosted in Europe on Frontière AI?
Partly. Scaleway (France) for Llama 3.3 70B: this one is labeled EU sovereign. Muse Spark 1.1 and Muse Spark 1.2 are served by US providers and labeled fast access — the API flags it before you call.
Is there a subscription or a minimum spend?
No subscription. You top up a prepaid balance from €10 and each call is debited at the per-token price. Calls are refused once the balance reaches zero; a call already running can take it slightly below — deducted from your next top-up.
Do I need to change my code to call Meta models?
No, if your code already uses an OpenAI SDK: point base_url at https://getfrontiereai.eu/api/v1, use a Frontière AI key and pass the model id, for example “llama-3.3-70b”. Streaming and tool calling are relayed as-is.
How are cached input tokens billed?
Muse Spark 1.1 and Muse Spark 1.2 have a discounted rate for prompt tokens the provider serves from its cache (€0.0252 per 100k tokens on Muse Spark 1.1). The provider reports these tokens on each call and we bill them at that rate; on a model without a cache rate, every input token is billed at the input rate.
Other providers
Per-token prices for every publisher in our catalog
All model prices on one page →Start calling Meta in minutes
Create an account, top up €10, change two lines of code.