AMD MI50 - 16GB usable
16GB of HBM2 at 1 TB/s, bandwidth that still embarrasses newer cards. Vega is off ROCm's supported list, so expect to pin old versions.
Specifications
| Brand | AMD |
|---|---|
| Model | MI50 |
| Usable VRAM | 16GB |
| Architecture | Vega |
| CUDA / Stream Processors | 3,840 |
| Memory Bandwidth | 1024 GB/s |
| TDP | 300W |
| FP32 TFLOPS | 13.3 |
Current Offers
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
US · USD · Weekly since May 2026 · Shipping may be extra
- eBay (no current offer)
- Amazon $618 current
- Newegg (no current offer)
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. Tax is excluded. Amazon shows a current price only.
Weekly offers table
| Date | eBay | Amazon | Newegg |
|---|---|---|---|
| Today, Nov 3, 2026 | – | $618.37 | – |
| Oct 9, 2026 | $199.00 | – | $432.99 |
| Oct 2, 2026 | $189.00 | – | $432.99 |
| Sep 25, 2026 | $189.00 | – | $432.99 |
| Sep 18, 2026 | $189.00 | – | $432.99 |
| Sep 11, 2026 | $172.50 | – | $432.99 |
| Sep 4, 2026 | $199.00 | – | $432.99 |
| Sep 1, 2026 | $173.00 | – | $432.99 |
| Aug 28, 2026 | $199.00 | – | $432.99 |
| Aug 21, 2026 | $199.00 | – | $432.99 |
| Aug 14, 2026 | $199.00 | – | $432.99 |
| Aug 7, 2026 | $198.00 | – | $432.99 |
| Jul 31, 2026 | $198.00 | – | $432.99 |
| Jul 24, 2026 | $198.00 | – | $432.99 |
| Jul 17, 2026 | $199.00 | – | $432.99 |
| Jul 10, 2026 | $179.95 | – | $432.99 |
| Jul 3, 2026 | $150.00 | – | $432.99 |
| Jun 26, 2026 | $199.00 | – | $432.99 |
| Jun 19, 2026 | $179.00 | – | $432.99 |
| Jun 12, 2026 | $199.00 | – | $432.99 |
| Jun 5, 2026 | $179.00 | – | $432.99 |
| May 29, 2026 | – | – | $432.99 |
| May 22, 2026 | – | – | $432.99 |
| May 15, 2026 | $149.99 | – | $432.99 |
| May 8, 2026 | $199.00 | – | $432.99 |
For AI / LLM Use
Good for 14B models. 30B requires aggressive quantization. Older architecture may have limited software support (check ROCm compatibility). Datacenter card with no display output, may need aftermarket cooling.
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: ~77 tokens/sec
Prefill: ~238 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
Cons
- 16GB usable VRAM: may need quantization for 30B+ models
- 300W TDP: high power draw
- Older Vega architecture: verify current ROCm support
- No display output: headless only
- May need aftermarket cooling solution
Community Verdict
No community reviews yet for the MI50. Know a good review? Let us know.