Meta API pricing
Tutujin API currently offers 20 Meta models (20 chat), pay-as-you-go with no monthly fee, callable through an OpenAI-compatible API.
Among token-billed models, the lowest input price is llama-3.1-8b-instruct at ¥0.2 per 1M tokens (Standard group).
Among per-request models, the cheapest is llama-3-sonar-large-32k-chat at ¥0.01 per request (Standard group).
Prices updated Nov 13, 2026, 6:08 PM
20
Models
2
Groups
1
Endpoints
¥0.2
Lowest input
/ 1M tokens
All Meta model prices
Each row shows the model's cheapest group, in CNY. Prices in other groups are on the model's own page.
Chat models20
| Model | Cheapest group | Input ¥/1M | Output ¥/1M | Per request ¥ |
|---|---|---|---|---|
| llama-3-sonar-large-32k-chat | Standard | — | — | ¥0.01 |
| llama-3-sonar-large-32k-online | Standard | — | — | ¥0.01 |
| llama-3.1-8b | Standard | — | — | ¥0.01 |
| llama-3.2-1b-instruct | Standard | — | — | ¥0.01 |
| code-llama-13b | Standard | — | — | ¥0.012 |
| code-llama-34b | Standard | — | — | ¥0.012 |
| code-llama-7b | Standard | — | — | ¥0.012 |
| llama-2-13b | Standard | — | — | ¥0.012 |
| llama-2-70b | Standard | — | — | ¥0.012 |
| llama-2-7b | Standard | — | — | ¥0.012 |
| llama-3-70b | Standard | — | — | ¥0.02 |
| llama-3-8b | Standard | — | — | ¥0.02 |
| llama-3-sonar-small-32k-chat | Standard | — | — | ¥0.02 |
| llama-3-sonar-small-32k-online | Standard | — | — | ¥0.02 |
| llama-3.1-405b | Standard | — | — | ¥0.02 |
| llama-3.2-11b-vision-instruct | Standard | — | — | ¥0.02 |
| llama-3.2-90b-vision-instruct | Standard | — | — | ¥0.04 |
| llama-3.1-70b | Standard | — | — | ¥0.1 |
| llama-3.1-8b-instruct | Standard | ¥0.2 | ¥0.3 | — |
| llama-3.2-3b-instruct | Standard | ¥0.2 | ¥0.2 | — |
How to call Meta models
One Tutujin API key is all you need; the example uses llama-3-sonar-large-32k-chat. Point base_url at https://www.tutujin.com/v1 and keep the rest of your official-SDK code.
curl https://www.tutujin.com/v1/chat/completions \
-H "Authorization: Bearer $TUTUJIN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "llama-3-sonar-large-32k-chat",
"messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]
}'FAQ
How is Meta API usage billed on Tutujin?
You pay for actual usage, with no monthly fee or minimum spend. Token-billed models charge input and output tokens separately; per-request models charge per call. The same model costs different amounts in different groups; this page shows each model's cheapest group. Prices updated Nov 13, 2026, 6:08 PM.
How do I call Meta models?
Create an API key in the console, set your SDK's base_url to https://www.tutujin.com/v1 and use a model name from the table on this page. Meta models on this site support these endpoints: Chat Completions.
Which groups can I use for Meta models?
Available groups: Standard, Economy (per call). A group decides the line a key uses and the price it pays; you choose one for each API key when you create it.
Can I get an invoice?
Yes. We issue both China domestic invoices (fapiao) and overseas invoices: add our WeChat support JsnonoChat and send your order number.