NVIDIA48 GB VRAM864 GB/s$0.7/hr

Llama 2 7B on L40S 48GB: Does It Fit, How Fast, What It Costs

Llama 2 7B at FP16 needs 13.5 GB for its weights and 47.8 GB in total at 8K context with 8 concurrent users. That means it fits on a single L40S 48GB. Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
47.8 GB needed / 48 GB available
Decode speed
41.66
tok/s per stream
Aggregate
499.9
tok/s @ 8 concurrent
Cost / 1M tokens
$0.3889
1× L40S 48GB
What changes as you add L40S 48GBs
Tensor-parallel degree vs VRAM, throughput, and cost. Cost scales linearly with GPU count; decode throughput scales near-linearly too, so cost per token stays roughly flat.
GPUsTotal VRAMVRAM neededDecode tok/sAggregate tok/sTTFTCost / 1M tokFits?
1× L40S48 GB47.8 GB41.66499.92.47 s$0.3889✓
2× L40S96 GB47.8 GB70.8217001.25 s$0.2288✓
4× L40S192 GB47.8 GB141.66799639.8 ms$0.1144✓
8× L40S384 GB47.8 GB283.327197334.9 ms$0.0572✓
Concurrency vs context length
On 1× L40S 48GB at FP16. KV cache scales linearly with context, so this is the number of users your GPUs can actually serve.
ContextConcurrent usersKV per requestFits?
4K162.15 GB✓ fits
8K84.29 GB✓ fits
33K217.18 GB✓ fits
66K134.36 GB✓ fits
131K068.72 GB✗ does not fit
Other GPUs that run Llama 2 7B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB2×3526$0.0788
AMD MI300X 192GB1×3067$0.2264
AMD MI325X 288GB1×3472$0.2800
A100 80GB SXM41×1157$0.2880
Intel Gaudi 3 128GB1×1909$0.2909
H100 SXM5 80GB1×1938$0.3583
Frequently asked questions

Can Llama 2 7B run on L40S 48GB?

Yes. Llama 2 7B at FP16 needs 47.8 GB including the KV cache, and L40S 48GB has 48 GB — so it fits on a single GPU.

How fast is Llama 2 7B on L40S 48GB?

41.66 tokens/sec per stream and 499.9 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 2.47 s, and inter-token latency is 24.0 ms.

What does Llama 2 7B on L40S 48GB cost per million tokens?

$0.3889 per 1M output tokens at 1× L40S 48GB ($0.7000/hr). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Llama 2 7B need on L40S 48GB?

13.5 GB for the weights at FP16, plus 34.4 GB of KV cache for 8 concurrent requests at 8K context — 47.8 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is L40S 48GB the right choice for Llama 2 7B?

It works. Compare $0.3889 per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 8× RTX 3080 10GB at $0.0186, and the fastest is 2× Cerebras CS-3 (WSE-3) at 39347 tok/s.

Adjust the workload yourself

Same formulas, your context length, concurrency, and quantisation.

Open in calculator