community-verified · measured pool data
Can the RTX PRO 6000 Blackwell 96GB ×2 run DeepSeek-V4-Flash-0731?
Yes — community-measured at 257.4 tok/s (2 runs at 8-bit).
Verified single-stream runs on the RTX PRO 6000 Blackwell 96GB ×2 from the localmaxxing public pool. Every number on this page is measured — nothing here is estimated.
| Quantization | Median tok/s | TTFT (ms) | Peak VRAM | Max context tested | Engines |
|---|---|---|---|---|---|
| 8-bit | 257.4 | 82.8 | 94.5 GB | 262,144 | vllm |