Mistral models and API pricing

Published Mistral models available through Meldport. Check capabilities before selecting a model for your integration.

Latest catalog evidence: Verified on 2026-11-06.

USD per million tokens. List prices are the catalog's published reference rates.
Model / publisher	Input	Output	List input / output	Published discount	Capabilities / context	Evidence
Mistral Large 4 Mistral	$0.61064	$1.87682	$1.36 / $4.18	Input: 55.1% Output: 55.1%	text, cache, tools, stream 524.29K tokens	Verified on 2026-11-06 Expires 2026-11-07

Select the exact model and variant

Compare the published context, capabilities and complete model ID before choosing a route. Prices and availability can change; inspect the evidence date on each entry.

One currently active example is Mistral Large 4, API ID meldport/mistral-large-4. Its published context is 524,288 tokens; published capabilities: text, cache, tools, stream. Evidence checked 2026-11-06. Review the model detail for its current evidence and limits.

Try a bounded Chat Completions request

Send this JSON to POST https://api.meldport.com/v1/chat/completions, with Content-Type: application/json and your dedicated key in the Authorization: Bearer header.

{
  "model": "meldport/mistral-large-4",
  "messages": [
    {
      "role": "user",
      "content": "Reply with one short sentence."
    }
  ],
  "max_tokens": 64,
  "stream": false
}

Store a dedicated key in your environment and follow the authentication quickstart. Use the exact ID above; a paid model ID and a :free alias have separate funding rules. See the eligible free aliases and shared monthly pool before assuming a free version is available.

Calculate ordinary input, cache and output costs, compare gateway tradeoffs, or read why your token ratio matters. Client-specific streaming and tool behavior must be tested separately; see agent compatibility.