Mistral-Small-3.1-24B

Advanced multimodal model with enhanced vision capabilities and extended context window

Model Details

An evolution of the Mistral Small series, combining advanced vision understanding with robust text processing capabilities. This model excels at both conversational and visual tasks while maintaining efficient performance. Optimized for practical deployment across a wide range of applications requiring multimodal understanding.

Parameters

24 billion

Context Window

128k tokens

Recommended Use

Ideal for multimodal applications, virtual assistants, and specialized domain fine-tuning requiring both text and vision capabilities.

Supported Languages

Multilingual with strong performance across major languages

Usage Examples

Installation:

pip install tinfoil

Inference:

from tinfoil import TinfoilAI

client = TinfoilAI(
    enclave="mistral-s-3-1-24b-p.model.tinfoil.sh",
    repo="tinfoilsh/confidential-mistral-small-3-1",
    api_key="YOUR_API_KEY",
)

chat_completion = client.chat.completions.create(
    messages=[
        {
            "role": "user",
            "content": "Hello!",
        }
    ],
    model="mistral-small-3-1-24b",
)
print(chat_completion.choices[0].message.content)

Tinfoil