GPU comparison
RTX PRO 6000 vs L40S
RTX PRO 6000 Blackwell doubles the L40S memory (96 vs 48 GB), almost doubles the bandwidth and adds FP4 Tensor Cores. The L40S remains the cheapest dedicated GPU for small models and classic rendering; for 30B–70B inference and heavy media work, the RTX PRO 6000 is worth the extra.
- Updated Nov 4, 2026
- Specs: vendor datasheets, dense Tensor figures
- Prices: live CoreGPU catalog
| Spec | NVIDIA RTX PRO 6000 Blackwell Server Edition | NVIDIA L40S |
|---|---|---|
| Architecture | Blackwell | Ada Lovelace |
| GPU memory | 96 GB GDDR7 | 48 GB GDDR6 |
| Memory bandwidth | 1.6 TB/s | 864 GB/s |
| FP8 dense | 1 PF | 0.73 PF |
| FP4 dense | 2 PF | n/a |
| Interconnect | PCIe 5.0 x16 | PCIe 4.0 x16 |
| Power (TDP) | 600 W | 350 W |
| Price per GPU / month | $1,199 | $849 |
| Price per GB of memory | $12.49 | $17.69 |
| Price per FP8 PFLOPS | $1,199 | $1,163 |
| Availability | 52 available | 38 available |
Green marks the better value on each line. Prices are per GPU per month at CoreGPU, at least 25% below the market median.
- Inference of 30B–70B models on a single card
- FP4 serving and video generation
- Large scenes and textures in rendering
$1,199 /GPU/moConfigure RTX PRO 6000
- 7B–13B models and diffusion at the lowest price
- Scale-out inference behind a load balancer
- Classic 3D rendering and VFX
$849 /GPU/moConfigure L40S