AMD RX 6950 XT - 16GB usable
Specifications
| Brand | AMD |
|---|---|
| Model | RX 6950 XT |
| Usable VRAM | 16GB |
| Architecture | RDNA2 |
| CUDA / Stream Processors | 5,120 |
| Memory Bandwidth | 576 GB/s |
| TDP | 335W |
| FP32 TFLOPS | 23.7 |
Current Offers
Used from €500
Prices last updated:
GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Price History
- eBay€500mid-range
For AI / LLM Use
Good for 14B models. 30B requires aggressive quantization. Verify ROCm support for the exact framework and operating system you plan to use.
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: ~43 tokens/sec
Prefill: ~423 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
- RDNA2 architecture: supported by current ROCm stacks
- Consumer card: easy to install, display output
Cons
- 16GB usable VRAM: may need quantization for 30B+ models
- Moderate memory bandwidth: not the fastest for inference
- 335W TDP: high power draw
- ROCm and CUDA-first framework compatibility should be checked before purchase
Community Verdict
No community reviews yet for the RX 6950 XT. Know a good review? Let us know.