Nvidia T600
Low-profile Turing professional card for multi-display and entry-level visualization.
- Arch
- Turing
- Memory
- 4 GB GDDR6
- Bandwidth
- 160 GB/s
- Released
- Q2 2021
At a glance
We don't currently track any T600 offers. See alternative GPUs.
Technical Specifications
Nvidia T600 · Per GPU
| Compute · dense | |
|---|---|
| FP32 | 1.7 TFLOPS |
| Precision support | FP16FP32INT8 |
| Memory | |
|---|---|
| Capacity | 4 GB GDDR6 |
| Bandwidth | 160 GB/s |
| Bus width | 128-bit |
| Silicon | |
|---|---|
| Architecture | Turing |
| Shader cores | 640 CUDA cores |
| Compute units | 10 SMs |
| Fabric and host | |
|---|---|
| Host interface | PCIe 3.0 x16 |
| Power | |
|---|---|
| Board power | 40 W |
| Cooling | Active |
Source: official Nvidia T600 datasheet.
Frequently Asked Questions
Why choose the T600?
4GB of GDDR6 at 40 W in a low-profile single-slot card with four Mini DisplayPort outputs. Suited to multi-display workstations, CAD viewing and light video editing in small chassis.
When is the T600 not a good fit?
No tensor cores and 4GB of VRAM, so it is not a card for AI inference or training. For tensor-core work or heavier visualization in the same low-profile form, the A2000 offers 12GB. Still suited to display output, CAD viewing and video playback.
Alternatives to Nvidia T600
-
-
Nvidia RTX A2000
12 GB GDDR6 · 80% more bandwidth than the T600
On-demand from $0.13 /GPU/hr
Last updated