Modelpublic

csrstka/Muse-Glimmer-30B-oQ4e-fp16

sourceHugging Faceupdated 1mo agoView on Hugging Face

0likes93downloads

Model Card

[!IMPORTANT] This quantization was uploaded on 2026-08-16 and replaces a previous version. If you downloaded this model before this date, please re-download for the updated weights.

Muse-Glimmer-30B-oQ4e-fp16

This model was quantized using oQ (oMLX v0.6.0) mixed-precision quantization.

Quantization details

—Model type: muse_glimmer
—Bits: 4
—Group size: 64
—Format: MLX safetensors