Transformers documentation

Mistral 3

Transformers

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v5.22.1).

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

이 모델은 2025-03-18에 Hugging Face Transformers에 기여되었습니다.

Mistral 3

Mistral 3는 포워드 패스당 시간을 줄이기 위해 레이어 수를 크게 줄인 지연 시간 최적화 모델입니다. 이 모델은 비전 이해 기능을 추가하고, 성능 저하 없이 최대 128K 토큰까지의 긴 컨텍스트 길이를 지원합니다.

원본 Mistral 3 체크포인트는 Mistral AI 조직에서 찾을 수 있습니다.

이 모델은 cyrilvallez와 yonigozlan이 기여했습니다. Mistral3를 다양한 작업에 적용하는 방법에 대한 추가 예제는 오른쪽 사이드바의 Mistral3 모델을 클릭하여 확인하세요.

use_kernels=True를 from_pretrained()에 설정하여 Hub의 최적화 커널로 지원되는 레이어를 대체합니다. 자세한 내용은 커널 로딩을 참고하세요.

아래 예제는 Pipeline와 AutoModel 클래스를 사용하여 이미지에 대한 텍스트를 생성하는 방법을 보여줍니다.

Pipeline

AutoModel

The input is a single word "Notes". 참고

Mistral 3은 텍스트 전용 생성을 지원합니다.

import torch
from transformers import AutoProcessor, AutoModelForImageTextToText

model_checkpoint = ".mistralai/Mistral-Small-3.1-24B-Instruct-2503"
processor = AutoProcessor.from_pretrained(model_checkpoint)
model = AutoModelForImageTextToText.from_pretrained(model_checkpoint, device_map="auto")

SYSTEM_PROMPT = "You are a conversational agent that always answers straight to the point, always end your accurate response with an ASCII drawing of a cat."
user_prompt = "Give me 5 non-formal ways to say 'See you later' in French."

messages = [
    {"role": "system", "content": SYSTEM_PROMPT},
    {"role": "user", "content": user_prompt},
]

text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = processor(text=text, return_tensors="pt").to(0)
generate_ids = model.generate(**inputs, max_new_tokens=50, do_sample=False)
decoded_output = processor.batch_decode(generate_ids[:, inputs["input_ids"].shape[1] :], skip_special_tokens=True)[0]

print(decoded_output)
"1. À plus tard!
 2. Salut, à plus!
 3. À toute!
 4. À la prochaine!
 5. Je me casse, à plus!

/_/\ ( o.o )

^ <

Mistral 3은 배치된 이미지 및 텍스트 입력을 지원합니다.

import torch
from transformers import AutoProcessor, AutoModelForImageTextToText

model_checkpoint = "mistralai/Mistral-Small-3.1-24B-Instruct-2503"
processor = AutoProcessor.from_pretrained(model_checkpoint)
model = AutoModelForImageTextToText.from_pretrained(model_checkpoint, device_map="auto")

messages = [
     [
         {
             "role": "user",
             "content": [
                 {"type": "image", "url": "https://huggingface.co/datasets/hf-internal-testing/transformers-synthetic-assets/resolve/main/images/llava_view.jpg"},
                 {"type": "text", "text": "Write a haiku for this image"},
             ],
         },
     ],
     [
         {
             "role": "user",
             "content": [
                 {"type": "image", "url": "https://www.ilankelman.org/stopsigns/australia.jpg"},
                 {"type": "text", "text": "Describe this image"},
             ],
         },
     ],
 ]


 inputs = processor.apply_chat_template(messages, padding=True, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt").to(model.device)

 output = model.generate(**inputs, max_new_tokens=25)

 decoded_outputs = processor.batch_decode(output, skip_special_tokens=True)
 decoded_outputs
["Write a haiku for this imageCalm waters reflect\nWhispers of the forest's breath\nPeace on wooden path"
, "Describe this imageThe image depicts a vibrant street scene in what appears to be a Chinatown district. The focal point is a traditional Chinese"]

Mistral 3는 또한 텍스트마다 서로 다른 수의 이미지를 가진 배치된 이미지 및 텍스트 입력을 지원했습니다. 아래 예제는 bitsandbytes로 모델을 양자화합니다.

import torch
from transformers import AutoProcessor, AutoModelForImageTextToText, BitsAndBytesConfig

model_checkpoint = "mistralai/Mistral-Small-3.1-24B-Instruct-2503"
processor = AutoProcessor.from_pretrained(model_checkpoint)
quantization_config = BitsAndBytesConfig(load_in_4bit=True)
model = AutoModelForImageTextToText.from_pretrained(
     model_checkpoint, quantization_config=quantization_config
 device_map="auto")

messages = [
     [
         {
             "role": "user",
             "content": [
                 {"type": "image", "url": "https://huggingface.co/datasets/hf-internal-testing/transformers-synthetic-assets/resolve/main/images/llava_view.jpg"},
                 {"type": "text", "text": "Write a haiku for this image"},
             ],
         },
     ],
     [
         {
             "role": "user",
             "content": [
                 {"type": "image", "url": "https://huggingface.co/datasets/hf-internal-testing/transformers-synthetic-assets/resolve/main/images/statue_of_liberty.jpg"},
                 {"type": "image", "url": "https://huggingface.co/datasets/hf-internal-testing/transformers-synthetic-assets/resolve/main/images/dreamstime_golden_gate_flowers.jpg"},
                 {"type": "text", "text": "These images depict two different landmarks. Can you identify them?"},
             ],
         },
     ],
 ]

 inputs = processor.apply_chat_template(messages, padding=True, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt").to(model.device)

 output = model.generate(**inputs, max_new_tokens=25)

 decoded_outputs = processor.batch_decode(output, skip_special_tokens=True)
 decoded_outputs
["Write a haiku for this imageSure, here is a haiku inspired by the image:\n\nCalm lake's wooden path\nSilent forest stands guard\n", "These images depict two different landmarks. Can you identify them? Certainly! The images depict two iconic landmarks:\n\n1. The first image shows the Statue of Liberty in New York City."]

Transformers

Mistral 3

The input is a single word "Notes". 참고

Mistral3Config

class transformers.Mistral3Config

MistralCommonBackend

클래스 transformers.MistralCommonBackend

add_special_tokens

add_tokens

apply_chat_template

batch_decode

build_inputs_with_special_tokens

convert_added_tokens

create_token_type_ids_from_sequences

decode

encode

from_pretrained

get_chat_template

get_special_tokens_mask

get_vocab

num_special_tokens_to_add

prepare_for_model

save_chat_templates

save_pretrained

save_vocabulary

토큰화

truncate_sequences

Mistral3Model

class transformers.Mistral3Model

forward

get_image_features

get_placeholder_mask

Mistral3ForConditionalGeneration

클래스 transformers.Mistral3ForConditionalGeneration

forward

get_image_features