community pool / model answer

GLM-5.3-Flash

chat · zai-org/GLM-5.3-Flash

148.4 tok/s · model median · 4 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
GB10 Grace Blackwell 256GB3-bit19.8 tok/sreported367105 GB1,000,000vllm
RTX PRO 6000 Blackwell 96GB ×23-bit171.9 tok/sreported177180.5 GB352,512vllm
RTX PRO 6000 Blackwell 96GB ×48-bit146.8 tok/sreported90- GB2,048vllm

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.