community pool / model answer

DeepSeek-V4-Flash-180B

chat · 0xSero/DeepSeek-V4-Flash-180B

40.2 tok/s · model median · 35 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Arc Pro B70 32GB ×48-bit41 tok/smeasured330- GB1,024vllm
GB10 Grace Blackwell 128GB8-bit23.3 tok/sreported2,481105.7 GB200,000vllm

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.