community pool / model answer

GLM-4.5-Air

chat · zai-org/GLM-4.5-Air

103 tok/s · model median · 3 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX PRO 6000 Blackwell 96GB4-bit108.9 tok/sreported32- GB32,768vllm
Ryzen AI Max 395 128GB4-bit6.2 tok/sreported1,131- GB783vllm
Ryzen AI Max 395 64GB3-bit29.6 tok/sreported3,74249.9 GB8,192llama.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.