Quantization-specific throughput and VRAM requirements for meta-llama/Meta-Llama-3-70B-Instruct running on NVIDIA H200 SXM 141GB.
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 | 34GB | 141GB | 249 tok/s | ✅ Fits |
| Q8 | 68GB | 141GB | 182 tok/s | ✅ Fits |
| FP16 | 137GB | 141GB | 94 tok/s | ✅ Fits |