isogen
/

Phi-4-reasoning-plus-exl3-4bpw

4-bit precision

Model card Files Files and versions

EXL3 quantization of Phi-4-reasoning-plus, 4 bits per weight.

Downloads last month: 6

Safetensors

Model size

4.12B 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/Phi-4-reasoning-plus-exl3-4bpw

Base model

microsoft/phi-4

Finetuned

microsoft/Phi-4-reasoning-plus

Quantized

(29)

this model