Quadro RTX 6000 vs Quadro RTX 8000
Manufacturer specifications of the Quadro RTX 6000 and the Quadro RTX 8000, side by side, per GPU. A value the manufacturer does not publish for the exact model is marked as such, never estimated.
NVIDIA Quadro RTX 6000 24GB
Turing
- GPU memory
- 24 GB
- Memory bandwidth
- 672 GB/s
Live price on EverXCloud · No offer right now
Suited workloads
NVIDIA Quadro RTX 8000 48GB
Turing
- GPU memory
- 48 GB
- Memory bandwidth
- 672 GB/s
Live price on EverXCloud · from $0.30 per GPU-hour
Suited workloads
Key differences
- The Quadro RTX 8000 has 2.0× the GPU memory of the Quadro RTX 6000 (48 GB vs 24 GB).
- Both have about the same memory bandwidth (672 vs 672 GB/s).
Specifications
| Specification | Quadro RTX 6000 | Quadro RTX 8000 |
|---|---|---|
| GPU memory | 24 GB | 48 GB (higher) |
| Memory type | GDDR6 | GDDR6 |
| Memory bandwidth | 672 GB/s | 672 GB/s |
| Max power | 295 W | 295 W |
| Architecture | Turing | Turing |
| Interconnect | NVLink bridge (100 GB/s) · PCIe 3.0 | NVLink bridge (100 GB/s) · PCIe 3.0 |
The higher value of each measured row is shown in bold. Manufacturer specifications, per GPU. Tensor throughput is shown without sparsity. Values the manufacturer does not publish for this exact model are left out.
What fits in memory, on one GPU
| Workload | Quadro RTX 6000 | Quadro RTX 8000 |
|---|---|---|
| LLM inference, 16-bit | ≈ 9B parameters | ≈ 15B parameters |
| LLM inference, 4-bit | ≈ 30B parameters | ≈ 60B parameters |
| Full training, mixed precision | ≈ 1B parameters | ≈ 2B parameters |
Cautious estimates counting 80% of memory for model weights; context length, batch size and framework change the real limit. Several GPUs together hold larger models.
Quadro RTX 6000 vs Quadro RTX 8000: frequently asked questions
Which has more memory, the Quadro RTX 6000 or the Quadro RTX 8000?
The Quadro RTX 8000 has 48 GB of GPU memory, against 24 GB for the Quadro RTX 6000.
Which is faster for LLM inference, the Quadro RTX 6000 or the Quadro RTX 8000?
Their memory bandwidth is about the same, so token generation speed is close for a model that fits on both; memory size decides which models fit. Measure your own model before a large run.
Can I rent the Quadro RTX 6000 and the Quadro RTX 8000 by the hour?
Yes, when sellers list them on the EverXCloud spot exchange: you buy GPU hours at the live price from a prepaid balance in euros, US dollars or pounds. Prices are shown excluding tax and EverXCloud is the merchant of record for every rental.
Be ready on launch day
Create your account now: you can top up your balance and rent the GPU that fits your workload as soon as the market opens.
Have GPU servers?
Businesses can join the EverXCloud Network and sell hours of their GPU servers on the spot exchange.