Mistral AI
Operacional· verificado a cada poucos minutos

Ministral 3 3B API

Preço de tabela de The Ministral 3 3B API: Mistral AI's smallest open-weight model, with vision, for fast, high-volume work, 70% below Mistral's

Preço por 1M tokens70%de desconto
Entrada
$0.03$0.10
Saída
$0.03$0.10
Obtenha a sua chave de APIExperimente no espaço de testes
ID do modeloministral-3-3b
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-3-3b",
    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="")

URL base https://api.zurelay.com/v1. Funciona com qualquer SDK da OpenAI. Documentação da API Chat

Janela de contexto
128K tokens
Raciocínio
Nenhum
Entrada
Text, Image
Lançamento
2/12/2025
Estado
Operacional

Preços

Preços da API Ministral 3 3B. Uma fração do preço de tabela.

Pague apenas o que utilizar com saldo pré-pago, sem subscrição. Carregue a partir de $10. Os pedidos que falham nunca são cobrados.

PreçoZurelayPreço de tabela de Mistral AI
Entrada
por 1M tokens
$0.03$0.10
Saída
por 1M tokens
$0.03$0.10

A mesma tarifa para qualquer comprimento de prompt, up to the full 128K context window. O streaming, as chamadas de ferramentas e as saídas estruturadas não têm custos adicionais.

Calculadora de poupança

Ajuste os controlos deslizantes ao seu consumo mensal.

Tokens de entrada por mês50M
Tokens de saída por mês10M
Zurelay
$1.80
Preço de tabela de Mistral AI
$6.00
Poupa todos os meses$4.20

$50.40 por ano

Visão geral

O que é Ministral 3 3B?

Ministral 3 3B is the smallest model in Mistral AI's Ministral 3 family, released on December 2, 2025: an open-weight model that reads text e images, made for fast, low-cost work such as classification, extraction, translation e image captions. The Zurelay Ministral 3 3B API serves its instruct version through one OpenAI-compatible endpoint with a 131,072-token context window, at 70% below Mistral's list price.

Mistral AI released the Ministral 3 family on December 2, 2025, alongside Mistral Large 3, in three sizes: 3B, 8B e 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 3B pairs a 3.4B-parameter language model with a 0.4B-parameter vision encoder, and Mistral says it fits in 8GB of GPU memory at FP8. On Zurelay the model ID is ministral-3-3b, and it custa $0.03 per 1M input tokens and $0.03 por 1M tokens de saída.

Mistral built Ministral 3 for edge e local use, e points the 3B model at lightweight, real-time work: image captions, text classification, translation, data extraction and short content. 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 56.8 on WildBench, level with Qwen3-VL-4B-Instruct e ahead of Gemma 3 4B (49.1), e 0.830 on MATH, ahead of Gemma 3 4B (0.759) and behind Qwen3-VL-4B-Instruct (0.900). Qwen3-VL-4B-Instruct also leads it on Arena Hard, 0.438 against 0.305. 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 e while streaming, e 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 360 tokens a second and knew little about events after 2024, as small models tend to, so put the facts it needs in the prompt. You pay $0.03 per 1M input tokens and $0.03 per 1M output tokens against Mistral's list price of $0.10 and $0.10.

Pontos fortes

Onde Ministral 3 3B se destaca. E o que as equipas criam com ele.

01

Fast e very cheap

Sobre 360 tokens a second in our tests, em $0.03 per 1M input tokens, for steps that run millions of times.

02

Lê imagens

A 0.4B-parameter vision encoder reads photos, screenshots e scans next to your text, for captions e quick visual checks.

03

Ferramentas e JSON

Chamada de funções e JSON output that Mistral describes as native, for routing, extraction e small agent steps.

04

Pesos abertos

Apache 2.0 weights em Hugging Face, small enough for 8GB de GPU memory, so what you build on Zurelay carries over to a local copy later.

Casos de uso

  • Classificação e encaminhamento

    Sort tickets, tag content e pick the next step in a pipeline, with JSON schema output for the code that follows.

  • Extração

    Pull fields from text, forms e screenshots into JSON em a very low price per call.

  • Translation

    Quick translation between the languages Mistral lists, including English, French, Spanish, German, Chinese, Japanese e Arabic.

  • Image captions

    Short descriptions e alt text for photos e screenshots at volume.

Começar

Faça chamadas a Ministral 3 3B em três passos.

Sem lista de espera nem um novo SDK. Se o seu código já comunica com a OpenAI, também comunica com a Zurelay.

  1. 1

    Criar uma chave

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

    Obtenha a sua chave de API
  2. 2

    Point your SDK em Zurelay

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

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

    Use ministral-3-3b

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

    ministral-3-3b

Funciona com as ferramentas que já usa

Comparar

Modelos semelhantes. Todos com a mesma chave.

70%de desconto

Ministral 3 14B

$0.06 / $0.06 por 1M tokens

Pick Ministral 3 14B, the largest Ministral 3 model, for stronger answers: 0.551 against 0.305 on Arena Hard in Mistral's tests, em twice the price per token.

API de Ministral 3 14B
53%de desconto

Qwen 3.8 Flash

$0.07 / $0.22 por 1M tokens

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

API de Qwen 3.8 Flash
90%de desconto

DeepSeek V4.1 Flash

$0.03 / $0.12 por 1M tokens

Pick DeepSeek V4.1 Flash for a 1M-token context e thinking on harder text and image tasks.

API de DeepSeek V4.1 Flash

Perguntas frequentes

Perguntas sobre a API de Ministral 3 3B.

Mais alguma dúvida? Fale com a nossa equipa e responderemos por email.

Quanto custa a API Ministral 3 3B?

On Zurelay, ministral-3-3b costs $0.03 per 1M input tokens e $0.03 per 1M output tokens. Mistral's list price is $0.10 input e $0.10 output per 1M, so you save 70%. One rate applies at every prompt length, and failed requests are never billed.

Como faço chamadas a the Ministral 3 3B API com o SDK da OpenAI?

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

O que é the Ministral 3 3B context window?

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

Can Ministral 3 3B read imagemms or PDFs?

It reads imagems: send them in image_url parts, as https links or base64 data URLs, e 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 3B: a request that includes one is turned away with a clear error before anything is billed.

Does Ministral 3 3B reason before it answers?

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

Does Ministral 3 3B support tool calling e 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, e tool_choice "none" works. Tool call ids must be at least 9 characters long, as the model's own e 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 3B or Ministral 3 14B?

Both read text e images, hold 131,072 tokens on Zurelay e call tools. 14B pairs a 13.5B-parameter language model with the same 0.4B vision encoder and scores higher in Mistral's tests (0.551 against 0.305 on Arena Hard, 68.5 against 56.8 on WildBench), at twice the price. 3B is faster, about 360 tokens a second against 170 in our tests, and suits simple, high-volume steps.

Is Ministral 3 3B open source?

Its weights are open under the Apache 2.0 license, which allows commercial use e changes. Mistral publishes base, instruct e reasoning versions on Hugging Face and says the 3B fits in 8GB 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 3B?

Yes. Requests for ministral-3-3b run on Mistral AI's Ministral 3 3B Instruct, e 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, e 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.

Mais modelos na Zurelay

Todos os modelos

Deixe de pagar o preço de tabela.
Comece a poupar hoje.

Crie uma chave em segundos, configure o SDK que já usa para aceder à Zurelay e cada pedido custa até 90% menos desde o primeiro token.

Obtenha a sua chave de APIFalar com um engenheiro