community pool / model answer

GLM-4.7-Flash

chat · zai-org/GLM-4.7-Flash

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

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Radeon AI Pro R9700 32GB ×24-bit92.9 tok/sreported110- GB783llama.cpp
Radeon AI Pro R9700 32GB ×34-bit97.4 tok/sreported25321.6 GB783llama.cpp
RTX 5090 32GB6-bit175.9 tok/sreported1325 GB8,192llama.cpp
RX 7700 XT 12GB4-bit21.5 tok/sreported102- 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.