NVIDIA RTX 3060 - 8GB usable
Specifications
| Brand | NVIDIA |
|---|---|
| Model | RTX 3060 |
| Usable VRAM | 8GB |
| Architecture | Ampere |
| CUDA / Stream Processors | 3,584 |
| Memory Bandwidth | 240 GB/s |
| TDP | 170W |
| FP32 TFLOPS | 13 |
Current Offers
No fresh used offer today · recently £167 on eBayNew from £380
Prices last updated:
GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Lowest tracked offers by retailer
UK · GBP · Weekly since Aug 2026 · Shipping may be extra
- eBay £380 at 90-day high
About these prices
Each line is the lowest offer we tracked from that retailer at weekly snapshots, ending at today's offer. The condition can vary between snapshots (new, used, or unrecorded for older ones). eBay prices exclude shipping.
Weekly offers table
| Date | eBay |
|---|---|
| Today, Nov 3, 2026 | £379.99 |
| Oct 30, 2026 | £379.99 |
| Oct 23, 2026 | £379.99 |
| Oct 9, 2026 | £167.10 |
| Oct 2, 2026 | £167.10 |
| Sep 25, 2026 | £167.10 |
| Sep 18, 2026 | £167.10 |
| Sep 11, 2026 | £167.10 |
| Sep 4, 2026 | £194.91 |
| Sep 1, 2026 | £146.30 |
| Aug 28, 2026 | £146.30 |
For AI / LLM Use
Limited VRAM restricts you to 7B quantized models. Slower generation, usable but not snappy.
What Models Can It Run?
- 7B Q6_K, 14B Q3_K (tight)
- 7B Q4_K_M only
Estimated Performance
Generation: ~18 tokens/sec
Prefill: ~232 tokens/sec
Recommended Quantisations
- Q4_K_M for 7B models
- Q3_K for larger experiments
Pros & Cons
Pros
- Ampere architecture: broad CUDA software support
- Consumer card: easy to install, display output
Cons
- Only 8GB usable VRAM: limited to small models
- Low memory bandwidth: slower token generation
Community Verdict
No community reviews yet for the RTX 3060. Know a good review? Let us know.