EXL3 quantization of Mistral-Small-3.1-24B-Instruct-2503, 3 bits per weight, no vision.

For vision and other bitrates: turboderp/Mistral-Small-3.1-24B-Instruct-2503-exl3.

Downloads last month
4
Safetensors
Model size
5.1B params
Tensor type
F16
·
I16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for isogen/Mistral-Small-3.1-24B-Instruct-2503-exl3-3bpw