NVIDIA RTX 5090 - 32GB usable
Specifications
| Brand | NVIDIA |
|---|---|
| Model | RTX 5090 |
| Usable VRAM | 32GB |
| Architecture | Blackwell |
| CUDA / Stream Processors | 21,760 |
| Memory Bandwidth | 1792 GB/s |
| TDP | 575W |
| FP32 TFLOPS | 105 |
Current Offers
Used from $4850New from $5150
Prices last updated:
GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Price History
- Amazon$4850current
- Newegg$5300mid-range
For AI / LLM Use
Solid choice for 30B models and comfortable 14B inference. Very fast generation speed.
What Models Can It Run?
- 30B Q6_K, 70B Q2_K
- 30B Q4_K_M, 14B full precision, 70B Q2 (tight)
- 14B Q6_K, 30B Q3_K (tight)
- 14B Q4_K_M, 7B full precision
- 7B Q6_K, 14B Q3_K (tight)
- 7B Q4_K_M only
Estimated Performance
Generation: ~134 tokens/sec
Prefill: ~1875 tokens/sec
Recommended Quantisations
- Q4_K_M recommended for 30B models
- Q6_K or Q8 for 14B and below
- Full precision for 7B
Pros & Cons
Pros
- 32GB usable VRAM: handles large models
- High memory bandwidth for fast generation
- Blackwell architecture: broad CUDA software support
- Consumer card: easy to install, display output
Cons
- 575W TDP: high power draw
Community Verdict
No community reviews yet for the RTX 5090. Know a good review? Let us know.