llmgauge

run it locally

Run Qwen2.5 72B on NVIDIA L40S

72.7B parameters on a 48 GB card. It fits at Q3_K_M — here's the VRAM math at every quantization, the speed to expect, and the quality you trade.

verdict · Qwen2.5 72B on NVIDIA L40S
Fits at Q3_K_M
needs 37.5 GB at Q3_K_M · 45.6 GB usable · ~23 tokens/sec
Q4_K_M won't fit here — run it in the cloud
Smallest GPU that fits at Q4_K_M: A100 80GB
Rent A100 80GB by the hour
VRAM needed and fit for Qwen2.5 72B on NVIDIA L40S by quantization.
quantization vram needed fits 48gb? tokens/sec quality
FP16 (full) 174.5 GB ✗ no — Reference quality — no quantization loss.
Q8_0 92.5 GB ✗ no — Near-lossless; rarely worth the extra space over Q6.
Q6_K 71.5 GB ✗ no — Virtually indistinguishable from full precision.
Q5_K_M 60.2 GB ✗ no — Minor loss; an excellent quality-vs-size balance.
Q4_K_M 48.9 GB ✗ no — Small but measurable loss; the popular default.
Q3_K_Mrecommended 37.5 GB ✓ yes ~23 Noticeable degradation; only when you're tight on VRAM.

Weights = params × bytes/weight, +20% for KV cache & runtime; usable VRAM is 95% of nameplate. Tokens/sec is a bandwidth ceiling (864 GB/s) — real throughput is lower with long context. Try other combinations →

Running Qwen2.5 72B on a NVIDIA L40S

At Q3_K_M, Qwen2.5 72B's weights take about 38 GB, inside the NVIDIA L40S's 46 GB of usable memory. Noticeable degradation; only when you're tight on VRAM. Generation speed is bound by memory bandwidth — the GPU reads the whole model once per token — so expect on the order of 23 tokens/sec before context overhead. To run it at the popular Q4_K_M default you'd want a A100 80GB or larger — or rent one by the hour rather than buy.

Quantization is the lever

Each step down in precision shrinks the model: FP16 needs about 174 GB, Q4_K_M about 49 GB — a 72% reduction for a small, usually acceptable quality cost. Q5_K_M and Q6_K are near-lossless if you have the headroom; drop to Q3 only when you're genuinely out of VRAM. The quantization guide covers the tradeoffs in detail.

Frequently asked questions

Can a NVIDIA L40S run Qwen2.5 72B?

Yes — Qwen2.5 72B fits on a NVIDIA L40S at Q3_K_M (38 GB of the 46 GB usable), with roughly 23 tokens/sec. Higher-precision quants need more VRAM; the table above shows each option.

How much VRAM does Qwen2.5 72B need?

Qwen2.5 72B is a 72.7B-parameter model. At FP16 that's about 174 GB; at Q4_K_M (the popular default) about 49 GB, including ~20% for the KV cache and runtime. Quantization is the main lever — see the per-quant table above.

Other combinations

Qwen2.5 72B on other GPUs: RTX 3090, RTX 4090, NVIDIA L4, RTX A6000, A100 40GB, A100 80GB

Other models on the NVIDIA L40S: Gemma 2 27B, Qwen2.5 32B, Command R 35B, Mixtral 8x7B, Llama 3.3 70B, Mixtral 8x22B

Related