NVIDIA RTX 5080 - 16GB
Specifications
| Brand | NVIDIA |
|---|---|
| Model | RTX 5080 |
| VRAM | 16GB |
| Architecture | Blackwell |
| CUDA / Stream Processors | 10,752 |
| Memory Bandwidth | 960 GB/s |
| TDP | 360W |
| FP32 TFLOPS | 56 |
Current Prices
GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Price History
Best price up 5% since 2026-03-13
Amazon
For AI / LLM Use
Good for 14B models. 30B requires aggressive quantization.
What Models Can It Run?
- 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: ~72 tokens/sec
Prefill: ~1000 tokens/sec
Recommended Quantisations
- Q4_K_M for 14B models
- Q6_K for 7B-8B models
- Q8 for 7B if VRAM allows
Pros & Cons
Pros
- High memory bandwidth for fast generation
- Blackwell architecture: good software support
- Consumer card: easy to install, display output
Cons
- 16GB VRAM: may need quantization for 30B+ models
- 360W TDP: high power draw
Community Verdict
No community reviews yet for the RTX 5080. Know a good review? Let us know.