NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Compare NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition against another machine in the Arena
Reference: Blackwell (microarchitecture) on Wikipedia
- Class
- GPU
- Memory
- 96 GB GDDR7
- Runs (Q4_K_M)
- —
- Bandwidth
- 1792 GB/s
- TDP
- 300 W
- Released
- 2025-03-18
- Price (US)
- $16528 new as of 2026-08
- Price (Canada)
- CA$23399 new sourced, as of 2026-08
Sourcing and disambiguation notes
Power-capped sibling of the Workstation Edition above -- same silicon, memory and capacity, TDP capped to 300W (NVIDIA states ~88% of the full card's AI throughput) for constrained-airflow multi-GPU builds. Shortage-pricing, ABOVE THE $15K SANITY THRESHOLD: gpupoet.com's tracked average of the three lowest-priced daily listings, against an $8,565-10,999 MSRP. Independently corroborated by a second source (Thunder Compute, 'last reviewed Aug 21, 2026') showing NVIDIA-marketplace-level pricing near $13k-16k for the RTX PRO 6000 family amid a 2026 shortage -- not a one-off scraping error, but genuinely volatile; same-day listings varied by thousands. No used/secondary-market figure found. Canadian price (2026-08, pre-tax): CA$23,399.00 for the PNY RTX PRO 6000 Blackwell Max-Q Edition, against the prior US $15,418 a 1.52x ratio, a normal band -- now a 1.42x ratio against the current $16,528. 2026-09 update: a price-refresh panel re-read the July page and proposed $10,999 -- verified directly against the page and rejected: that figure is the card's static 2025 launch MSRP, explicitly labelled as such on the page, not a tracked price; every dated figure on the page (June $15,282, July $15,418, current live snapshot $21,789) sits far above MSRP and trending upward, the opposite direction from the panel's proposed value. Replaced instead with the page's own August 2026 dated section (same lowest-average-of-three methodology): $16,528, range $14,299-$20,009.
Specification sources
- https://www.nvidia.com/en-us/products/workstations/professional-desktop-gpus/rtx-pro-6000-max-q/ — vendor, 2025-03-18
Measurements
Decode is token generation — the speed you feel while an answer streams. Prefill is prompt processing — the wait before it starts. Why bandwidth predicts decode speed.
No records yet for this hardware. Know of a published benchmark on it? Submit the link.
What verified, single-source and estimated mean, and the same rows with every filter and sort in the benchmarks explorer.
Speed over time
What it can run
Fit is arithmetic, not a measurement — how it is computed. Against 96 GB; models that fit are listed largest first.
No modeled quant fits in 96 GB.