Meta Llama models, priced in toman
8 Meta Llama models are live on uttapen. Prices are in toman per 1 million tokens, and you are charged for the tokens you actually use.
Meta Llama on uttapen
Meta Llama currently has 8 models in the uttapen catalogue. All of them bill against a toman wallet, for the tokens a request actually used.
The spread inside Meta Llama alone is wide: Llama 3.1 8B Instruct runs about 59 toman for a 1,000-word request, while Llama 4 Maverick costs 608 toman for the same one. Picking the right model inside a brand matters as much as picking the brand.
Measured against the whole catalogue, the median Meta Llama output price undercuts about 82% of the other paid models.
On capabilities: 3 take image input, 4 support tool calling. The longest context belongs to Llama 4 Maverick, at 524,288 tokens.
Every Meta Llama model answers on the same sk-up- key and the same base_url: switching between them means changing the `model` string and nothing else.
Which Meta Llama model should I start with?
Every Meta Llama model, priced in toman
| Model | Capabilities | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|
Llama 3.1 70B Instructmeta-llama/llama-3.1-70b-instruct | 131,072 | 249,278 | 249,278 | |
Llama 3.1 8B Instructmeta-llama/llama-3.1-8b-instruct | Tool callingJSON output | 131,072 | 17,311 | 27,698 |
Llama 3.2 1B Instructmeta-llama/llama-3.2-1b-instruct | 60,000 | 9,348 | 69,590 | |
Llama 3.2 3B Instructmeta-llama/llama-3.2-3b-instruct | JSON output | 131,072 | 17,623 | 115,983 |
Llama 3.3 70B Instructmeta-llama/llama-3.3-70b-instruct | Tool callingJSON output | 131,072 | 69,244 | 180,034 |
Llama 4 Maverickmeta-llama/llama-4-maverick | Image inputTool callingJSON output | 524,288 | 121,177 | 346,219 |
Llama 4 Scoutmeta-llama/llama-4-scout | Image inputTool callingJSON output | 327,680 | 34,622 | 103,866 |
Llama Guard 4 12Bmeta-llama/llama-guard-4-12b | Image input | 163,840 | 62,319 | 62,319 |
Common questions about the Meta Llama API
- How do I reach the Meta Llama API from Iran?
- Sign up with your mobile number, top the wallet up in toman and create an sk-up- key. Then, in the official OpenAI SDK, point base_url at https:/
/ api.uttapen.ir/ v1 and set model to the Meta Llama model id. No international card, no VPN. - What do Meta Llama models cost in toman?
- It depends on the model: the cheapest, Llama 3.1 8B Instruct, is about 59 toman for a 1,000-word request, and the dearest, Llama 4 Maverick, about 608 toman. Each model page carries its own exact price, and you are billed for what the request actually used.
- How many Meta Llama models are on uttapen?
- 8. 4 support tool calling and 3 accept image input. The catalogue re-syncs hourly, so a newly released model shows up here on its own.