NVIDIA L40S~34 tok/s (Q4)

Qwen/Qwen2.5-Math-72B-Instruct speed on NVIDIA L40S

Quantization-specific throughput and VRAM requirements for Qwen/Qwen2.5-Math-72B-Instruct running on NVIDIA L40S.

Speed Snapshot

Topline estimate from compatibility data

ModelQwen/Qwen2.5-Math-72B-Instruct

GPUNVIDIA L40S

Q4 speed34 tok/s

Q4 VRAM required36GB

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	36GB	48GB	34 tok/s	✅ Fits
Q8	71GB	48GB	21 tok/s	❌ Not recommended
FP16	142GB	48GB	13 tok/s	❌ Not recommended