Mistral Small 3.1 is a 24B parameter multimodal LLM that excels in a wide variety of tasks, including text generation, instruction following, and image understanding. It features a large 128,000 token context window and can be deployed locally, fitting within a single RTX 4090 or a 32GB RAM MacBook once quantized[1][5].

Some noteworthy features of Mistral Small 3.1 include state-of-the-art vision capabilities, advanced reasoning, and strong performance in programming and mathematical tasks[1][8].

Metric	Value
Parameter Count	24 billion
Mixture of Experts	No
Context Length	128,000 tokens
Multilingual	Yes
Quantized*	Unknown

*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.