mistral-large-4:cloud

336 Downloads Updated 18 hours ago

Mistral Large 4 is Mistral's open-weight, general-purpose multimodal model. Its granular Mixture-of-Experts design activates 52B of 1.05T total parameters, with a 1.6B vision encoder handling images.

vision tools thinking cloud

Cost /1M tokens

$0.68

input

$0.07

cached

$2.09

output

Context

1M tokens

Size

1.05T parameters

ollama run mistral-large-4:cloud

curl http://localhost:11434/api/chat \
  -d '{
    "model": "mistral-large-4:cloud",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='mistral-large-4:cloud',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'mistral-large-4:cloud',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Readme

Mistral Large 4 is a natively multimodal model with 1 trillion parameters, 49 billion of them active. It’s Mistral’s largest and most capable model to date, and it’s still improving as they refine it.

It’s strong across coding, agentic workflows, and multimodal understanding. It’s competitive with the strongest open-source models, and well ahead of open-weight models from the US and Europe. On enterprise workloads like cybersecurity, finance, and law, Mistral reports state-of-the-art results among open models.

Reference

https://mistral.ai/news/mistral-large-4/