NVIDIA141 GB VRAM4,800 GB/s$4/hr

Llama 3 8B on H200 SXM5 141GB: Does It Fit, How Fast, What It Costs

Llama 3 8B at FP16 needs 16.1 GB for its weights and 24.6 GB in total at 8K context with 8 concurrent users. That means it fits on a single H200 SXM5 141GB. Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
24.6 GB needed / 141 GB available
Decode speed
194.3
tok/s per stream
Aggregate
2331
tok/s @ 8 concurrent
Cost / 1M tokens
$0.4766
1× H200 SXM5 141GB
What changes as you add H200 SXM5 141GBs
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× H200141 GB24.6 GB194.32331300.9 ms$0.4766✓
2× H200282 GB24.6 GB330.37926165.9 ms$0.2804✓
4× H200564 GB24.6 GB660.53170598.0 ms$0.1402✓
8× H2001128 GB24.6 GB132112682064.0 ms$0.0701✓
Concurrency vs context length
On 1× H200 SXM5 141GB at FP16. KV cache scales linearly with context, so this is the number of users your GPUs can actually serve.
ContextConcurrent usersKV per requestFits?
4K232549.8 MB✓ fits
8K1161.07 GB✓ fits
33K294.29 GB✓ fits
66K148.59 GB✓ fits
131K717.18 GB✓ fits
Other GPUs that run Llama 3 8B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB1×870.3$0.1596
AMD MI300X 192GB1×2574$0.2698
V100 SXM2 32GB1×437.1$0.3177
A100 40GB SXM41×752.8$0.3321
AMD MI325X 288GB1×2914$0.3336
A100 80GB SXM41×971.4$0.3432
Frequently asked questions

Can Llama 3 8B run on H200 SXM5 141GB?

Yes. Llama 3 8B at FP16 needs 24.6 GB including the KV cache, and H200 SXM5 141GB has 141 GB — so it fits on a single GPU.

How fast is Llama 3 8B on H200 SXM5 141GB?

194.3 tokens/sec per stream and 2331 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 300.9 ms, and inter-token latency is 5.1 ms.

What does Llama 3 8B on H200 SXM5 141GB cost per million tokens?

$0.4766 per 1M output tokens at 1× H200 SXM5 141GB ($4.00/hr). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Llama 3 8B need on H200 SXM5 141GB?

16.1 GB for the weights at FP16, plus 8.6 GB of KV cache for 8 concurrent requests at 8K context — 24.6 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is H200 SXM5 141GB the right choice for Llama 3 8B?

It works. Compare $0.4766 per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 4× RTX 3080 10GB at $0.0443, and the fastest is 1× Cerebras CS-3 (WSE-3) at 9714 tok/s.

Adjust the workload yourself

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

Open in calculator