Back to Quant Hub

Solar 10.7B Instruct

11B

Upstage

Depth-upscaled 10.7B punching above weight. Strong on reasoning benchmarks.

Consumer GPUMac / Apple Silicon

4K

Max Context

2

Quant Variants

GGUF Q4_K_M

Best Quality

97.0%

Accuracy Retained

Calculate VRAMHugging FaceCompare

Quantization Variants

Per-quant VRAM, quality loss, and inference speed on RTX 4090

FormatLevelBPWVRAMPPL LossSpeedActions
GGUFQ4_K_M4.857.2 GB3.0%125 tok/s
CalcHF
AWQINT446.5 GB4.2%168 tok/s
CalcHF