Quantization-specific throughput and VRAM requirements for google-bert/bert-base-uncased running on AMD Instinct MI210.
Speed values come from the compatibility dataset (`estimatedTokensPerSec`) and are sorted by quantization.
For full verdict logic and alternate GPUs, see the canonical compatibility page.
Open full compatibility report| Quantization | VRAM needed | VRAM available | Speed | Verdict |
|---|---|---|---|---|
| Q4 | 1GB | 64GB | 276 tok/s | ✅ Fits |
| Q8 | 1GB | 64GB | 192 tok/s | ✅ Fits |
| FP16 | 1GB | 64GB | 112 tok/s | ✅ Fits |