Distillation-Aware Weight Quant of models. The RTN is further trained using data generated from BF16 model or 8bit RTN quant if BF16 is not feasible.
-
hermitdave/Agnes-3.0-Flash-DWQ-4bit-MTP
Text Generation • 32B • Updated • 91 -
hermitdave/LFM2.5-24B-A2B-MLX-DWQ-3bit-v2
Text Generation • 24B • Updated • 392 -
hermitdave/LFM2.5-24B-A2B-MLX-DWQ-3bit
Text Generation • 24B • Updated • 2.92k -
hermitdave/LFM2.5-24B-A2B-MLX-DWQ-4bit
Text Generation • 24B • Updated • 208