Mistral AI
稼働中· 数分ごとに確認

Ministral 3 14B API

The Ministral 3 14B API: the largest Ministral 3 model, with visionとtool calling, 70% below Mistral'sの定価

1Mトークンあたりの料金70%割引
入力
$0.06$0.20
出力
$0.06$0.20
APIキーを取得Playgroundで試す
モデルIDministral-3-14b
chat.py
from openai import OpenAI
client = OpenAI(
    base_url="https://api.zurelay.com/v1",
    api_key="YOUR_ZURELAY_KEY",
)
stream = client.chat.completions.create(
    model="ministral-3-14b",
    messages=[{"role": "user", "content": "Explain quicksort in two sentences."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices:
        print(chunk.choices[0].delta.content or "", end="")

ベースURL https://api.zurelay.com/v1. どのOpenAI SDKにも対応しています。 チャットAPIドキュメント

コンテキストウィンドウ
128K トークン
推論
なし
入力
Text, Image
リリース日
2025年12月2日
ステータス
稼働中

料金

Ministral 3 14BのAPI料金。 定価より大幅にお得です。

プリペイド残高からの従量課金で、サブスクリプションは不要です。$10からチャージできます。失敗したリクエストには課金されません。

料金ZurelayMistral AIの定価
入力
1M トークンあたり
$0.06$0.20
出力
1M トークンあたり
$0.06$0.20

プロンプトの長さにかかわらず同じ料金です, up to the full 128K context window。ストリーミング、ツール呼び出し、構造化出力に追加料金はかかりません。

節約額の計算

スライダーで月間使用量を設定してください。

月間入力トークン数50M
月間出力トークン数10M
Zurelay
$3.60
Mistral AIの定価
$12.00
毎月の節約額$8.40

年間$100.80

概要

Ministral 3 14Bとは？

70% below Mistral's list price.でMinistral 3 14B is the largest model in Mistral AI's Ministral 3 family, released on December 2, 2025: an open-weight model that reads textとimages and, by Mistral's account, performs comparably to its larger Mistral Small 3.2 24B. The Zurelay Ministral 3 14B API serves its instruct version through one OpenAI-compatible endpoint with a 131,072-token context window,

Mistral AI released the Ministral 3 family on December 2, 2025, alongside Mistral Large 3, in three sizes: 3B, 8Bと14B. Each size comes as a base, an instruct and a reasoning model, all with image understanding and all under the Apache 2.0 license, with the weights on Hugging Face. Ministral 3 14B pairs a 13.5B-parameter language model with a 0.4B-parameter vision encoder, and Mistral says it fits in 24GB of GPU memory at FP8. On Zurelay the model ID is ministral-3-14b, and itの出力料金は1Mトークンあたり$0.06 per 1M input tokens and $0.06です。

Mistral built Ministral 3 for edgeとlocal use,とpoints the 14B model at private chat and assistant deployments on limited hardware, local agents and fine-tuning. It handles dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. Mistral says its Ministral instruct models often write an order of magnitude fewer tokens than comparable models.

On Mistral's model card the instruct model scores 0.551 on Arena Hard, 68.5 on WildBenchと0.904 on MATH, ahead of Qwen3 14B without thinking (0.427, 65.1と0.870) and Gemma 3 12B (0.436, 63.2 and 0.854). On MM MTBench, a multimodal chat test, it scores 8.49 against 6.70 for Gemma 3 12B. These are Mistral's numbers, not independent ones.

The instruct model answers directly without reasoning first, so reasoning_effort has no effect. It calls tools, several at onceとwhile streaming,とreturns JSON for json_object or json_schema in response_format. Mistral lists a 256K context window on its own API; on Zurelay a request holds up to 131,072 tokens. In our tests it wrote about 170 tokens a second and knew little about events after 2024, so put the facts it needs in the prompt. You pay $0.06 per 1M input tokens and $0.06 per 1M output tokens against Mistral's list price of $0.20 and $0.20.

強み

Ministral 3 14Bが得意なこと。 チームでの活用例も紹介します。

01

Strong for its size

0.551 on Arena Hardと68.5 on WildBench in Mistral's tests, ahead of Qwen3 14BとGemma 3 12B, and comparable to Mistral Small 3.2 24B by Mistral's account.

02

画像を読み取り

MM MTBench in Mistral's tests. Send screenshots, chartsとphotos next to your text.上の8.49

03

ツールとJSON

Calls several tools at once, also while streaming,とfollows a JSON schema, for agent stepsとextraction.

04

公開ウェイト

GPU memory件中FP8, so what you build on Zurelay carries over to a self-hosted copy later.でApache 2.0 weights on Hugging Face that fit in 24GB件

ユースケース

  • Assistantsとchat

    a very low price per token.でCustomer-facing chatとinternal assistants that need good everyday answers

  • Agent steps

    Tool-calling steps such as looking things up, filling formsとchoosing the next action, with JSON for the code that follows.

  • Extraction from text and 枚の画像

    Pull fields from emails, documentsとscreenshots into a JSON schema.

  • Multilingual work

    Translate, summarizeとanswer in the languages Mistral lists, from FrenchとGerman to Chinese, Japanese and Arabic.

始める

3ステップでMinistral 3 14Bを呼び出せます。

ウェイトリストも新しいSDKも不要です。OpenAIに対応したコードなら、そのままZurelayでも使えます。

  1. 1

    キーを作成

    Sign up, add creditとcreate an API key. Set a monthly budget or a rate limit per key if you like.

    APIキーを取得
  2. 2

    ZurelayでPoint your SDK

    Change the base URL. Everything else in your code stays the same.

    https://api.zurelay.com/v1
  3. 3

    Use ministral-3-14b

    Send chat completions as usual. Streaming, tool callsとusage reporting work as you expect.

    ministral-3-14b

普段使っているツールに対応

比較

類似モデル。 すべて同じキーで使えます。

70%割引

Ministral 3 3B

1Mトークンあたり入力$0.03 / 出力$0.03

half the price per token, about twice as fast in our tests.でPick Ministral 3 3B for the simplest, highest-volume steps

Ministral 3 3B API
53%割引

Qwen 3.8 Flash

1Mトークンあたり入力$0.07 / 出力$0.22

Pick Qwen 3.8 Flash for a 1M-token context, video inputとa thinking mode when a task needs more than a small model.

Qwen 3.8 Flash API
61%割引

Gemini 3.1 Flash-Lite

1Mトークンあたり入力$0.10 / 出力$0.59

Pick Gemini 3.1 Flash-Lite for Google's low-cost model with a 1M-token contextとaudio, video and PDF input.

Gemini 3.1 Flash-Lite API

よくある質問

Ministral 3 14B APIについてよくある質問。

ほかにご質問はありますか？ チームに問い合わせると、メールでお答えします。

Ministral 3 14B APIの料金はいくらですか？

On Zurelay, ministral-3-14b costs $0.06 per 1M input tokensと$0.06 per 1M output tokens. Mistral's list price is $0.20 inputと$0.20 output per 1M, so you save 70%. One rate applies at every prompt length, and failed requests are never billed.

OpenAI SDKでthe Ministral 3 14B APIを呼び出すには？

Install the official openai package, set base_url to https://api.zurelay.com/v1とuse your Zurelay API key. Then send a Chat Completions request with model set to "ministral-3-14b". Zurelay also accepts ministral-3-14b-instruct, mistralai/ministral-3-14bとministral-14b. Mistral recommends a temperature of 0.1 or lower for production work.

Ministral 3 14B context windowとは？

On Zurelay a request holds up to 131,072 tokens. Mistral lists a 256K window on its own API. The model reads textとimagesとwrites text.

画像Can Ministral 3 14B read枚s or PDFs?

画像It reads枚s: send them in image_url parts, as https links or base64 data URLs,とthey count as input tokens. Mistral suggests images close to square, with very wide or thin ones cropped. Zurelay doesn't take PDFs, audio or video for Ministral 3 14B: a request that includes one is turned away with a clear error before anything is billed.

Does Ministral 3 14B reason before it answers?

No. Zurelay serves the instruct model, which answers directly, so reasoning_effort has no effectとthere are no reasoning tokens～pay for. Mistral also publishes a separate Ministral 3 14B Reasoning model, which Zurelay doesn't serve. For answers with thinking, use Qwen 3.8 Flash or DeepSeek V4.1 Flash.

Does Ministral 3 14B support tool callingとJSON output?

Yes. Pass tools as you would to any OpenAI model; it can call several at once, also while streaming. A tool forced by name is called,とtool_choice "none" works. Tool call ids must be at least 9 characters long, as the model's ownとOpenAI's are. For JSON, send response_format with json_object or json_schema and the answer comes back as JSON.

Should I use Ministral 3 14B or Ministral 3 3B?

Both read textとimages, hold 131,072 tokens on Zurelayとcall tools. 14B is the stronger model, with a 13.5B-parameter language model against 3.4B, and scores higher in Mistral's tests: 0.551 against 0.305 on Arena Hard and 68.5 against 56.8 on WildBench. 3B costs half as much and was about twice as fast in our tests, so it suits the simplest, highest-volume steps.

Is Ministral 3 14B open source?

Its weights are open under the Apache 2.0 license, which allows commercial useとchanges. Mistral publishes base, instructとreasoning versions on Hugging Face and says the 14B fits in 24GB of GPU memory at FP8. The API is the quick way to use it without running your own hardware.

Is this the real Ministral 3 14B?

Yes. Requests for ministral-3-14b run on Mistral AI's Ministral 3 14B Instruct,とZurelay doesn't swap in a different model. Only smart routing, which you turn on yourself, can answer with a close alternative when the model is down,とthe x-zurelay-model header names the model that answered. Responses are sampled, so wording varies between runs. Zurelay is independent and is not affiliated with or endorsed by Mistral AI.

Zurelayのその他のモデル

すべてのモデル

定価で払うのはもう終わり。
今日から節約を始めましょう。

数秒でキーを作成し、お使いのSDKをZurelayに接続するだけ。最初のトークンから、すべてのリクエストが最大90%安くなります。

APIキーを取得エンジニアに相談