The NVIDIA Quadro K620 is an entry-level workstation GPU from the Maxwell generation. Its certified professional drivers made it a favourite for CAD viewers, multi-monitor workstations and remote desktops. In 2026 it is available mainly as a low-cost GPU VPS.
In this 2026 guide we cover the Quadro K620’s specs, what it can (and cannot) do today, where you can still rent it, and which GPU to choose instead if you need more power for AI.
Table of Contents
Quadro K620 Specifications
| Spec | Quadro K620 |
|---|---|
| Architecture | Maxwell (GM107), 2014 |
| CUDA cores | 384 |
| Memory | 2 GB DDR3, 128-bit |
| TDP | 45 W |
| Video encoder (NVENC) | Yes (1st-gen Maxwell) |
| Driver status | Maintenance (Maxwell moved to legacy after R580) |
Is the Quadro K620 Still Worth Renting in 2026?
For light professional work, remote desktops and 2D CAD, the K620 is still usable and very cheap. NVIDIA has moved Maxwell GPUs to maintenance-only drivers, so do not expect new features.
It is not suitable for AI, 3D rendering or gaming. For those, choose a newer card with more VRAM.
What it is good for
- Remote Windows desktops and virtual workstations
- 2D CAD and viewing lightweight 3D models
- Multi-monitor office work and video playback
AI workloads on the Quadro K620
Too little VRAM for modern AI models. Fine for remote desktops, light video playback and legacy apps.
Where to Rent the Quadro K620
1. GPU Mart
- Dedicated GPU servers and GPU VPS in US data centers
- Express GPU VPS: GT 730 / P600 / K620 (2 GB) from $21/mo
- Windows or Linux with full admin access
- K620 GPU VPS from $21/mo · T1000 dedicated $99/mo
Pros
- Very low monthly prices
- Dedicated hardware, no noisy neighbours
- One-click AI apps (Ollama, Stable Diffusion)
Cons
- Monthly billing only
- US locations only
GPU Mart’s Express GPU VPS ($21/month) includes K620-class GPUs. It is the cheapest way to get a professional GPU in the cloud.
For heavier CAD or 3D work, the Quadro T1000 (8 GB) dedicated server at $99/month is a much stronger workstation card.
Quadro K620 vs Alternatives (Monthly Prices)
If your workload needs more speed or VRAM, these are the closest upgrades available as hosted servers today:
| GPU | VRAM | From | Notes |
|---|---|---|---|
| Quadro P1000 | 4 GB | $37/mo | Cheapest dedicated GPU at GPU Mart |
| GTX 1650 | 4 GB | $65.45/mo | NVENC Turing encoder, emulators |
| RTX 5060 (VPS) | 8 GB | $85/mo | Modern Blackwell GPU, good for small AI models |
| Quadro T1000 | 8 GB | $99/mo | Quiet workstation card, CAD |
| RTX A4000 (VPS) | 16 GB | $119/mo | Best value for Stable Diffusion and 14B LLMs |
How to Choose
- Need AI? Pick at least 8 GB of VRAM, ideally 16–24 GB. Older low-VRAM cards cannot run modern AI models well.
- Need video encoding? Choose a card with a modern NVENC encoder (Turing or newer, such as GTX 1650, T1000 or RTX cards).
- Need Windows apps or emulators? Any dedicated GPU with Windows support works; more VRAM helps with multiple instances.
- Budget. A $21–$99/month plan is fine for light work; for AI, the RTX A4000 VPS at $119/month is the best step up.
FAQ
GPU Mart offers K620-class GPUs in its Express GPU VPS from $21/month.
For 2D drafting and light 3D, yes. For large 3D models, choose a Quadro T1000 or RTX A4000.
No. 2 GB of memory is too little for modern AI models.
Conclusion
The Quadro K620 remains a cheap, reliable card for remote desktops and 2D CAD. GPU Mart offers it from $21/month. For 3D, rendering or AI, step up to a T1000 or RTX A4000.
Prices were checked in September 2026 and change often. Always confirm current pricing on the provider’s website before you order.