Mistral models, priced in toman
23 Mistral models are live on uttapen. Prices are in toman per 1 million tokens, and you are charged for the tokens you actually use.
Mistral on uttapen
Mistral is the 5th largest of the 65 suppliers in the uttapen catalogue, with 23 models. All of them bill against a toman wallet, for the tokens a request actually used.
The spread inside Mistral alone is wide: Mistral Nemo runs about 27 toman for a 1,000-word request, while Mistral Medium 3.5 costs 4,051 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 Mistral output price undercuts about 66% of the other paid models.
On capabilities: 12 take image input, 20 support tool calling, 4 do multi-step reasoning, 1 accepts audio. The longest context belongs to Mistral Large 4, at 1,048,576 tokens.
Every Mistral 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 Mistral model should I start with?
- Mistral Nemo27 · per 1,000 wordsThe cheapest way in: about 27 toman for a 1,000-word request.
- Mistral Large 41,260 · per 1,000 wordsFor multi-step problems, maths and debugging: it reasons before it answers.
- Ministral 3 14B 2512180 · per 1,000 wordsWhen the input is an image or a scanned invoice.
- Mistral Embed 231245.01 · per 1,000 wordsAn embedding model for semantic search and RAG; only input tokens are billed.
Every Mistral model, priced in toman
| Model | Capabilities | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|
Mistral Largemistralai/mistral-large | Tool callingJSON outputFile input | 128,000 | 692,439 | 2,077,316 |
Codestral 2508mistralai/codestral-2508 | Tool callingJSON outputFile input | 256,000 | 103,866 | 311,597 |
Devstral 2 2512mistralai/devstral-2512 | Tool callingJSON outputFile input | 262,144 | 138,488 | 692,439 |
Ministral 3 14B 2512mistralai/ministral-14b-2512 | Image inputTool callingJSON output | 262,144 | 69,244 | 69,244 |
Ministral 3 3B 2512mistralai/ministral-3b-2512 | Image inputTool callingJSON output | 131,072 | 34,622 | 34,622 |
Ministral 3 8B 2512mistralai/ministral-8b-2512 | Image inputTool callingJSON output | 262,144 | 51,933 | 51,933 |
Mistral Large 2407mistralai/mistral-large-2407 | Tool callingJSON outputFile input | 131,072 | 692,439 | 2,077,316 |
Mistral Large 3 2512mistralai/mistral-large-2512 | Image inputTool callingJSON outputFile input | 262,144 | 173,110 | 519,329 |
Mistral Large 4mistralai/mistral-large-4-0 | Image inputTool callingReasoningJSON output | 1,048,576 | 242,354 | 727,061 |
Mistral Medium 3mistralai/mistral-medium-3 | Image inputTool callingJSON outputFile input | 131,072 | 138,488 | 692,439 |
Mistral Medium 3.5mistralai/mistral-medium-3-5 | Image inputTool callingReasoningJSON output | 262,144 | 519,329 | 2,596,645 |
Mistral Small 4mistralai/mistral-small-2603 | Image inputTool callingReasoningJSON output | 262,144 | 51,933 | 207,732 |
Voxtral Small 24B 2507mistralai/voxtral-small-24b-2507 | Tool callingJSON outputFile inputAudio input | 32,768 | 34,622 | 103,866 |
Mistral Medium 3.1mistralai/mistral-medium-3.1 | Image inputTool callingJSON outputFile input | 131,072 | 138,488 | 692,439 |
Mistral Small 3.2 24Bmistralai/mistral-small-3.2-24b-instruct | Image inputTool callingJSON output | 256,000 | 25,966 | 69,244 |
Mistral Small 3.1 24Bmistralai/mistral-small-3.1-24b-instruct | Image inputTool callingJSON output | 128,000 | 121,523 | 192,152 |
Sabamistralai/mistral-saba | Tool callingJSON outputFile input | 32,768 | 69,244 | 207,732 |
Mistral Small 3mistralai/mistral-small-24b-instruct-2501 | JSON output | 32,768 | 17,311 | 27,698 |
Mistral Nemomistralai/mistral-nemo | Tool callingJSON output | 131,072 | 10,040 | 10,387 |
Mixtral 8x22B Instructmistralai/mixtral-8x22b-instruct | Tool callingJSON outputFile input | 65,536 | 692,439 | 2,077,316 |
Mistral Embed 2312mistralai/mistral-embed-2312 | Embeddings | 8,192 | 34,622 | — |
Codestral Embed 2505mistralai/codestral-embed-2505 | Embeddings | 8,192 | 51,933 | — |
Mistral Medium 3.5Batchmistralai/mistral-medium-3-5:batch | Image inputTool callingReasoningJSON output | 262,144 | 259,665 | 1,298,323 |
Common questions about the Mistral API
- How do I reach the Mistral 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 Mistral model id. No international card, no VPN. - What do Mistral models cost in toman?
- It depends on the model: the cheapest, Mistral Nemo, is about 27 toman for a 1,000-word request, and the dearest, Mistral Medium 3.5, about 4,051 toman. Each model page carries its own exact price, and you are billed for what the request actually used.
- How many Mistral models are on uttapen?
- 23. 20 support tool calling and 12 accept image input. 2 are embedding models for /v1/
embeddings. The catalogue re-syncs hourly, so a newly released model shows up here on its own.