MistralHosted on Luminet

Mistral Large 3

mistral/mistral-large-3

Try in Playground

Mistral's flagship refresh. 200B dense, 256K context, multilingual, multimodal. Apache 2.0.

Context window

256Ktokens

Input price

$1.80/ 1M tokens

Output price

$5.40/ 1M tokens

P50 latency

165ms

Quick start

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.luminet.ai/v1",
  apiKey: process.env.LUMINET_API_KEY,
});

const resp = await client.chat.completions.create({
  model: "mistral/mistral-large-3",
  messages: [{ role: "user", content: "Hello!" }],
});

Capabilities

textvisioncode

Throughput: 230 tok/s · Released 2026-04