Inference Net models, priced in toman
2 Inference Net models are live on uttapen. Prices are in toman per 1 million tokens, and you are charged for the tokens you actually use.
Inference Net on uttapen
Inference Net currently has 2 models in the uttapen catalogue. All of them bill against a toman wallet, for the tokens a request actually used.
The spread inside Inference Net alone is wide: Schematron V2 Turbo runs about 81 toman for a 1,000-word request, while Schematron V2 Small costs 126 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 Inference Net output price undercuts about 88% of the other paid models.
The longest context among Inference Net models is Schematron V2 Small, at 128,000 tokens.
Every Inference Net 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 Inference Net model should I start with?
Every Inference Net model, priced in toman
| Model | Capabilities | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|
Schematron V2 Smallinference-net/schematron-v2-small | JSON output | 128,000 | 17,311 | 79,630 |
Schematron V2 Turboinference-net/schematron-v2-turbo | JSON output | 128,000 | 10,387 | 51,933 |
Common questions about the Inference Net API
- How do I reach the Inference Net 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 Inference Net model id. No international card, no VPN. - What do Inference Net models cost in toman?
- It depends on the model: the cheapest, Schematron V2 Turbo, is about 81 toman for a 1,000-word request, and the dearest, Schematron V2 Small, about 126 toman. Each model page carries its own exact price, and you are billed for what the request actually used.
- How many Inference Net models are on uttapen?
- 2. The catalogue re-syncs hourly, so a newly released model shows up here on its own.