Modelpublic

dfp-official/DeepSeek-V4-Flash-0731-oQ8-mtp

sourceHugging Faceupdated 2mo agoView on Hugging Face

0likes176downloads

Model Card

DeepSeek-V4-Flash-0731-oQ8-mtp

This model was quantized using oQ (oMLX v0.5.3) mixed-precision quantization.

Quantization details

—Model type: deepseek_v4
—Bits: 8
—Group size: 64
—Format: MLX safetensors