NVIDIA RTX PRO 6000 GPU: Specs, VRAM, Price & Benchmarks (2026)
The NVIDIA RTX PRO 6000 (96GB GDDR7 with ECC) launched in 2025. This guide covers its full specs, VRAM, SXM/PCIe differences where they apply, AI performance, and live per-GPU rental pricing on Aquanode.
Short answer on cost: the RTX PRO 6000 rents from $1.38 per GPU per hour on Aquanode.
How much VRAM does the RTX PRO 6000 have?
The RTX PRO 6000 has 96GB GDDR7 with ECC, with 1,792 GB/s of peak memory bandwidth.
- VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs RTX A6000 48GB GDDR6 with ECC
- VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs L40S 48GB GDDR6 with ECC
What fits in 96GB GDDR7 with ECC of VRAM
| Model | Precision | Fits? |
|---|---|---|
| Llama 3.1 70B | FP16 | ~140GB. Does not fit on one 96GB card; needs 2. |
| Llama 3.1 70B | INT4 | ~35-40GB. Fits comfortably with plenty of headroom. |
| Mixtral 8x22B | INT4 | ~70GB. Fits on a single card. |
| Llama 3 8B | FP16 | ~16GB. Fits easily. |
Approximate, based on published parameter counts and standard bytes-per-parameter rules of thumb (FP16 ≈ 2 bytes/param, INT4 ≈ 0.5-0.6 bytes/param). Real footprint also depends on KV-cache size and framework overhead.
What can the RTX PRO 6000 run?
Popular open models from small to frontier scale, with the memory each needs and how many RTX PRO 6000 cards (96GB GDDR7 with ECC each) that takes.
| Model | As published | FP8 | INT4 |
|---|---|---|---|
| Qwen/Qwen3-8B 8.2B | BF16: ~18.3 GB, 1 GPU | FP8: ~9.2 GB, 1 GPU | INT4: ~4.6 GB, 1 GPU |
| Qwen/Qwen2.5-14B-Instruct 14.8B | BF16: ~33 GB, 1 GPU | FP8: ~16.5 GB, 1 GPU | INT4: ~8.3 GB, 1 GPU |
| Qwen/Qwen3-32B 32.8B | BF16: ~73.2 GB, 1 GPU | FP8: ~36.6 GB, 1 GPU | INT4: ~18.3 GB, 1 GPU |
| Qwen/Qwen-72B 72.3B | BF16: ~162 GB, 2 GPUs | FP8: ~80.8 GB, 1 GPU | INT4: ~40.4 GB, 1 GPU |
| MiniMaxAI/MiniMax-M2.7 228.7B | FP8: ~256 GB, 3 GPUs | – | INT4: ~128 GB, 2 GPUs |
| deepseek-ai/DeepSeek-R1 684.5B | FP8: ~765 GB, 8 GPUs | – | INT4: ~383 GB, 4 GPUs |
Estimates: weights at the stated precision plus a flat 20% for KV cache and overhead, at a moderate context length. A dash means the precision is not offered for that model (it is already published at that size). INT4 needs a published quantized checkpoint. Open any model for a per-GPU breakdown, or use the RTX PRO 6000 VRAM calculator.
RTX PRO 6000 VRAM calculator: check which models fit in its memory at each precision.
All models that fit in 96 GB: the open models whose weights and overhead fit, at native, FP8 and INT4 precision.
RTX PRO 6000 specs
| Architecture | NVIDIA, launched 2025 |
| VRAM | 96GB GDDR7 with ECC |
| Memory bandwidth | 1,792 GB/s |
| TDP | 600W |
| Form factor | PCIe, dual-slot |
Specs sourced from the vendor's public datasheet/product page. See the source.
Related reading: RTX PRO 6000 Blackwell guide, RTX PRO 6000 vs RTX 5090, and NVIDIA DGX Spark guide.
GPU Glossary: What is VRAM?, Tensor Cores, CUDA Cores, TFLOPS
RTX PRO 6000 AI performance
96GB of Blackwell-generation VRAM in a workstation card. The largest single-card capacity outside the data-center accelerators, enough to hold a 70B model at FP8 (about 70GB of weights) on one card without splitting across GPUs.
- Memory bandwidth: 1,792 GB/s
No NVLink, so scaling past one card means falling back to slower PCIe-only communication, and at 600W it draws meaningfully more power than the older-generation workstation cards (A6000, L40S) it competes with.
The cheapest RTX PRO 6000 offer right now ($1.38/GPU/hr) is about 37% below the market median of $2.19/GPU/hr.
RTX PRO 6000 price: what does it cost?
Buying. $8,565 at launch (March 2025) (launch MSRP). Reseller pricing had risen to roughly $13,000-$16,000 by September 2026 amid a GDDR7/VRAM component shortage; check current reseller pricing rather than assuming the launch figure still holds.
Renting. The cheapest current on-demand rate for the RTX PRO 6000 on Aquanode is $1.98/GPU/hr. Live rates range from $1.38 to $2.41 per GPU per hour, with a median of $2.19/GPU/hr. Billed by the offer's own terms; the table below shows every live rate.
| Region | $/GPU/hr | Available | VRAM | vCPU | RAM |
|---|---|---|---|---|---|
| United States | $1.38 | 1 | 95 GB | 48 | 240 GB |
| Zrh, Ch | $1.66 | 1 | 96 GB | 506 | 258 GB |
| Spain | $1.84 | 1 | 96 GB | 16 | 125 GB |
| United States | $1.98 | 2 | 96 GB | 24 | 218 GB |
| Chennai, In | $2.40 | 1 | 96 GB | 28 | 160 GB |
| Beltsville, Us | $2.41 | 1 | 96 GB | 16 | 144 GB |
| Finland | $2.91 | 1 | 96 GB | 30 | 90 GB |
RTX PRO 6000 price history
Aquanode stores one snapshot of its GPU price index per UTC day. For the RTX PRO 6000 that is 26 days so far, 2026-10-06 to 2026-10-31, so this is a short history, not a long-run trend. The lowest per-GPU rate went from $1.53 on 2026-10-06 to $1.38 on 2026-10-31.
| Day (UTC) | Lowest per GPU hour | Median per GPU hour | Data-center lowest per GPU hour | Offers |
|---|---|---|---|---|
| 2026-10-31 | $1.38 | $2.19 | $1.98 | 74 |
| 2026-10-30 | $1.66 | $2.19 | $1.98 | 50 |
| 2026-10-29 | $1.66 | $2.19 | $1.98 | 47 |
| 2026-10-28 | $1.56 | $2.20 | $1.98 | 38 |
| 2026-10-27 | $1.66 | $2.17 | $1.98 | 34 |
| 2026-10-26 | $1.66 | $2.19 | $1.98 | 40 |
| 2026-10-25 | $1.61 | $2.19 | $1.98 | 44 |
| 2026-10-24 | $1.58 | $2.20 | $1.98 | 39 |
| 2026-10-23 | $1.61 | $2.19 | $1.98 | 42 |
| 2026-10-22 | $1.66 | $2.41 | $1.98 | 39 |
| 2026-10-21 | $1.61 | $2.13 | $1.98 | 44 |
| 2026-10-20 | $1.66 | $2.20 | $1.98 | 40 |
| 2026-10-19 | $1.51 | $2.19 | $1.98 | 48 |
| 2026-10-18 | $1.12 | $2.19 | $1.98 | 49 |
| 2026-10-17 | $1.66 | $2.19 | $1.98 | 45 |
| 2026-10-16 | $1.21 | $2.74 | $1.98 | 89 |
| 2026-10-15 | $1.66 | $2.59 | $1.98 | 87 |
| 2026-10-14 | $1.21 | $2.46 | $1.98 | 122 |
| 2026-10-13 | $1.32 | $2.43 | $1.98 | 114 |
| 2026-10-12 | $1.38 | $2.38 | $1.98 | 126 |
| 2026-10-11 | $1.38 | $2.41 | $1.98 | 139 |
| 2026-10-10 | $1.47 | $2.36 | $1.98 | 143 |
| 2026-10-09 | $1.29 | $2.19 | $1.98 | 122 |
| 2026-10-08 | $1.38 | $2.11 | $1.98 | 105 |
| 2026-10-07 | $1.54 | $2.13 | $1.98 | 111 |
| 2026-10-06 | $1.53 | $2.02 | $1.98 | 76 |
Each row is the stored daily snapshot of the live index, copied as recorded. A dash means that day stored no figure. The same series is available as JSON and summarised in the monthly GPU price report.
How this price is calculated
All prices on this page are normalized to a per-GPU hourly rate using each offer's authoritative GPU count, so that raw price is divided by the number of GPUs it actually covers; some offers report price as already per-GPU, so those are used as-listed. An offer with a missing, zero, or invalid GPU count is excluded entirely rather than published at a guessed rate.
No offers were excluded from this snapshot for a missing or invalid price. No offers were dropped as price outliers in this snapshot.
Only the cheapest qualifying offer per provider is shown in the table above. This page regenerates at most once per hour.
RTX PRO 6000 vs RTX A6000: how do they compare?
- VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs RTX A6000 48GB GDDR6 with ECC
- Memory bandwidth: 1,792 GB/s vs 768 GB/s
On Aquanode right now, RTX A6000 starts at $0.550/GPU/hr against RTX PRO 6000's $1.38/GPU/hr, about 60% less.
RTX PRO 6000 vs L40S: how do they compare?
- VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs L40S 48GB GDDR6 with ECC
- Memory bandwidth: 1,792 GB/s vs 864 GB/s
On Aquanode right now, L40S starts at $0.809/GPU/hr against RTX PRO 6000's $1.38/GPU/hr, about 41% less.
Compare the RTX PRO 6000 with other GPUs
Compare RTX PRO 6000 with
- B200 vs RTX PRO 6000
- H200 vs RTX PRO 6000
- H100 vs RTX PRO 6000
- A100 vs RTX PRO 6000
- DGX A100 vs RTX PRO 6000
- RTX PRO 6000 vs V100
- AMD MI300X vs RTX PRO 6000
- L40S vs RTX PRO 6000
- L40 vs RTX PRO 6000
- A40 vs RTX PRO 6000
- RTX A6000 vs RTX PRO 6000
- RTX A5000 vs RTX PRO 6000
- RTX A4000 vs RTX PRO 6000
- L4 vs RTX PRO 6000
- RTX PRO 6000 vs T4
- RTX 6000 Ada vs RTX PRO 6000
- RTX PRO 6000 vs RTX PRO 6000 WS
- RTX PRO 6000 vs RTX PRO 6000 SE
- RTX PRO 5000 vs RTX PRO 6000
- RTX 5090 vs RTX PRO 6000
- RTX 5080 vs RTX PRO 6000
- RTX 5070 Ti vs RTX PRO 6000
- RTX 5070 vs RTX PRO 6000
- RTX 5060 Ti vs RTX PRO 6000
- RTX 4090 vs RTX PRO 6000
- RTX 4080 Super vs RTX PRO 6000
- RTX 4080 vs RTX PRO 6000
- RTX 4070 Ti vs RTX PRO 6000
- RTX 4070 Super vs RTX PRO 6000
- RTX 4070 vs RTX PRO 6000
- RTX 4060 Ti vs RTX PRO 6000
- RTX 3090 vs RTX PRO 6000
- RTX 3080 vs RTX PRO 6000
- RTX 3070 vs RTX PRO 6000
- RTX 3060 vs RTX PRO 6000
- B300 vs RTX PRO 6000
- AMD MI355X vs RTX PRO 6000
- AMD MI325X vs RTX PRO 6000
- H100 NVL vs RTX PRO 6000
- RTX PRO 4500 vs RTX PRO 6000
- RTX PRO 4500 SE vs RTX PRO 6000
- RTX PRO 4000 vs RTX PRO 6000
- RTX 5880 Ada vs RTX PRO 6000
- RTX 5000 Ada vs RTX PRO 6000
- RTX 4000 Ada vs RTX PRO 6000
- RTX 4000 SFF Ada vs RTX PRO 6000
- RTX 2000 Ada vs RTX PRO 6000
- RTX 4070 Ti Super vs RTX PRO 6000
- RTX A4500 vs RTX PRO 6000
- RTX 2080 Ti vs RTX PRO 6000
- RTX 3080 Ti vs RTX PRO 6000
- RTX 3070 Ti vs RTX PRO 6000
- RTX 3060 Ti vs RTX PRO 6000
- GTX 1080 Ti vs RTX PRO 6000
- RTX 4060 vs RTX PRO 6000
- RTX 5060 vs RTX PRO 6000
- GTX 1660 Super vs RTX PRO 6000
- RTX 2060 vs RTX PRO 6000
- Radeon RX 9070 XT vs RTX PRO 6000
- Radeon RX 6700 XT vs RTX PRO 6000
- RTX 6000 vs RTX PRO 6000
- A16 vs RTX PRO 6000
- P40 vs RTX PRO 6000
- P4 vs RTX PRO 6000
- Quadro P2000 vs RTX PRO 6000
- Quadro M4000 vs RTX PRO 6000
Get notified when the price drops
GPU supply moves hourly. Tell us what you're waiting for and we'll email you when a matching offer appears across any provider we track.
Rent a RTX PRO 6000 on Aquanode
- On-demand instances from $1.38/GPU/hr, billed by the provider's own terms, with no hardware procurement or long-term commitment.
- 74 live offers across 6 regions today.
- Set a price/availability alert above to hear the moment a cheaper or newly-available RTX PRO 6000 offer appears.
- Compare every RTX PRO 6000 offer side by side, or browse the full multi-provider GPU marketplace.
Good for
96GB of Blackwell-generation VRAM in a workstation card. The largest single-card capacity outside the data-center accelerators, enough to hold a 70B model at FP8 (about 70GB of weights) on one card without splitting across GPUs.
Not good for
No NVLink, so scaling past one card means falling back to slower PCIe-only communication, and at 600W it draws meaningfully more power than the older-generation workstation cards (A6000, L40S) it competes with.
RTX PRO 6000 FAQs
How much VRAM does the RTX PRO 6000 have?
The RTX PRO 6000 has 96GB GDDR7 with ECC, with 1,792 GB/s of peak memory bandwidth.
What is the NVIDIA RTX PRO 6000?
The NVIDIA RTX PRO 6000 is a GPU released in 2025, with 96GB GDDR7 with ECC of memory and a 600W power envelope. See the full spec table above for interconnect, form factor and tensor-throughput details.
How much does it cost to rent a RTX PRO 6000?
Live RTX PRO 6000 rental prices currently range from $1.38 to $2.41 per GPU per hour, with a median of $2.19 per GPU per hour.
What's the cheapest RTX PRO 6000 rate?
The lowest current RTX PRO 6000 rate on Aquanode is $1.38 per GPU per hour in United States.
How does the RTX PRO 6000 compare to the RTX A6000?
VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs RTX A6000 48GB GDDR6 with ECC Memory bandwidth: 1,792 GB/s vs 768 GB/s On Aquanode right now, RTX A6000 starts at $0.550/GPU/hr against RTX PRO 6000's $1.38/GPU/hr, about 60% less. See the full RTX PRO 6000 vs RTX A6000 comparison for a shared-provider price breakdown.
How does the RTX PRO 6000 compare to the L40S?
VRAM: RTX PRO 6000 96GB GDDR7 with ECC vs L40S 48GB GDDR6 with ECC Memory bandwidth: 1,792 GB/s vs 864 GB/s On Aquanode right now, L40S starts at $0.809/GPU/hr against RTX PRO 6000's $1.38/GPU/hr, about 41% less. See the full RTX PRO 6000 vs L40S comparison for a shared-provider price breakdown.
What is the RTX PRO 6000 good for?
96GB of Blackwell-generation VRAM in a workstation card. The largest single-card capacity outside the data-center accelerators, enough to hold a 70B model at FP8 (about 70GB of weights) on one card without splitting across GPUs.
What are the RTX PRO 6000's limitations?
No NVLink, so scaling past one card means falling back to slower PCIe-only communication, and at 600W it draws meaningfully more power than the older-generation workstation cards (A6000, L40S) it competes with.
Is renting cheaper than buying?
Renting avoids the upfront hardware cost and lets you match spend to actual usage. A rented RTX PRO 6000 at $1.38/hr only costs money while it's running, whereas buying ties up capital in hardware that keeps depreciating whether it's in use or not. Buying outright runs $8,565 at launch (March 2025) (launch MSRP). Which is cheaper depends on how continuously you'd run it; short or bursty workloads usually favor renting.
How is the RTX PRO 6000 price calculated?
All prices are normalized to a per-GPU hourly rate using each offer's authoritative GPU count, which the raw price is divided by; some offers report price as already per-GPU. Offers whose price can't be safely normalized, or whose rate is an extreme outlier against the rest of the market, are excluded.
RTX PRO 6000 price by region
Related guides
Other models in the same generation, then the rest of the GPU index.
