community pool / model answer

GLM-Z1-32B-0414

chat · zai-org/GLM-Z1-32B-0414

16.4 tok/s · model median · 17 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 3060 12GB ×24-bit15.5 tok/smeasured7,24519.8 GB65,536llama.cpp
RTX 3090 24GB4-bit21.6 tok/smeasured5,986- GB131,072llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.