NVIDIA H100 SXM5 80GB~190 tok/s (Q4)

RedHatAI/Llama-3.3-70B-Instruct-FP8-dynamic speed on NVIDIA H100 SXM5 80GB

Quantization-specific throughput and VRAM requirements for RedHatAI/Llama-3.3-70B-Instruct-FP8-dynamic running on NVIDIA H100 SXM5 80GB.

Speed Snapshot

Topline estimate from compatibility data

ModelRedHatAI/Llama-3.3-70B-Instruct-FP8-dynamic

GPUNVIDIA H100 SXM5 80GB

Q4 speed190 tok/s

Q4 VRAM required34GB

Data Source

Calculation and benchmark status

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.

Quantization Speed Table

Quantization	VRAM needed	VRAM available	Speed	Verdict
Q4	34GB	80GB	190 tok/s	✅ Fits
Q8	68GB	80GB	123 tok/s	✅ Fits
FP16	137GB	80GB	66 tok/s	❌ Not recommended