Pixtral 12B is a Multimodal LLM that excels in handling both images and text, supporting tasks like image captioning, visual question answering, and document analysis. It maintains strong performance in text-only tasks as well.

Metric	Value
Parameter Count	12 billion
Mixture of Experts	No
Context Length	128,000 tokens
Multilingual	Unknown
Quantized*	Yes
Precision*	bf16

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