AMD MI210 - 64GB usable

64GB of HBM2e on one card, by far the easiest route to that capacity. ROCm only, so plan for the software side.

Specifications

BrandAMD
ModelMI210
Usable VRAM64GB
ArchitectureCDNA2
CUDA / Stream Processors6,656
Memory Bandwidth1638 GB/s
TDP300W
FP32 TFLOPS23

Current Offers

Used from £4200

Prices last updated:

GPUDojo is reader-supported. When you buy through links on our site, we may earn an affiliate commission.

Price History

  • eBay£4200mid-range
Aug 21Aug 28Sep 1Sep 4£4200£4200

For AI / LLM Use

Best for running 70B+ models on a single GPU. Very fast generation speed. Datacenter card with no display output, may need aftermarket cooling.

What Models Can It Run?

  • 70B Q4_K_M, 30B full precision
  • 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: ~123 tokens/sec

Prefill: ~411 tokens/sec

Recommended Quantisations

  • Q6_K or Q8 for best quality
  • Q4_K_M for larger models
  • Full precision for 30B and below

Pros & Cons

Pros

  • 64GB usable VRAM: handles large models
  • High memory bandwidth for fast generation
  • CDNA2 architecture: supported by current ROCm stacks

Cons

  • 300W TDP: high power draw
  • ROCm and CUDA-first framework compatibility should be checked before purchase
  • No display output: headless only
  • May need aftermarket cooling solution

Community Verdict

No community reviews yet for the MI210. Know a good review? Let us know.