community pool / model answer

DeepSeek-V4.1-Flash

chat · deepseek-ai/DeepSeek-V4.1-Flash

72 tok/s · model median · 13 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
GB10 Grace Blackwell 512GB4-bit59.1 tok/smeasured391- GB430,080vllm
GB10 Grace Blackwell 512GB8-bit31.7 tok/sreported3,328- GB2,048vllm
RTX PRO 6000 Blackwell Max-Q Workstation Edition 96GB ×48-bit303.7 tok/smeasured221- GB524,288vllm

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.