community-verified · measured pool data
Can the RTX PRO 6000 Blackwell 96GB run DeepSeek-V4-Flash?
Yes — community-measured at 63.9 tok/s (2 runs at 3-bit).
Verified single-stream runs on the RTX PRO 6000 Blackwell 96GB 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 |
|---|---|---|---|---|---|
| 3-bit | 63.9 | 5699 | 93.9 GB | 524,288 | llama.cpp |