NVIDIA32 GB VRAM1,792 GB/s$0.5/hr

Llama 3.3 70B on RTX 5090 32GB: Does It Fit, How Fast, What It Costs

Llama 3.3 70B at FP8 (E4M3) needs 70.6 GB for its weights and 92.1 GB in total at 8K context with 8 concurrent users. That means it needs 4× RTX 5090 32GB. Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 4×
92.1 GB needed / 128 GB available
Decode speed
84.14
tok/s per stream
Aggregate
4039
tok/s @ 8 concurrent
Cost / 1M tokens
$0.1376
4× RTX 5090 32GB

RTX 5090 32GB has no NVLink, so 4-way tensor parallelism runs over PCIe. tokcalc applies a flat 0.85× multi-GPU efficiency factor for this, which is realistic for NVLink-equipped datacenter cards but optimistic for PCIe scale-out — treat the throughput and cost figures for this configuration as an upper bound, and measure before committing.

What changes as you add RTX 5090 32GBs
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× RTX32 GB92.1 GB0.0000.000——✗
2× RTX64 GB92.1 GB0.0000.000——✗
4× RTX128 GB92.1 GB84.1440395.55 s$0.1376✓
8× RTX256 GB92.1 GB168.3161552.79 s$0.0688✓
Concurrency vs context length
On 4× RTX 5090 32GB at FP8 (E4M3). KV cache scales linearly with context, so this is the number of users your GPUs can actually serve.
ContextConcurrent usersKV per requestFits?
4K421.34 GB✓ fits
8K212.68 GB✓ fits
33K510.74 GB✓ fits
66K221.47 GB✓ fits
131K142.95 GB✓ fits
Other GPUs that run Llama 3.3 70B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
A100 80GB SXM42×1127$0.5916
H100 SXM5 80GB2×1888$0.7358
AMD MI300X 192GB1×878.3$0.7906
L40S 48GB2×486.8$0.7988
AMD MI325X 288GB1×994.3$0.9778
Intel Gaudi 3 128GB1×546.9$1.02
Frequently asked questions

Can Llama 3.3 70B run on RTX 5090 32GB?

Yes, but it needs 4× RTX 5090 32GB. At FP8 (E4M3) the total footprint is 92.1 GB (70.6 GB of weights plus 21.5 GB of KV cache at 8K × 8 concurrent users), against 128 GB of available VRAM across 4 GPUs.

How fast is Llama 3.3 70B on RTX 5090 32GB?

84.14 tokens/sec per stream and 4039 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 5.55 s, and inter-token latency is 11.9 ms.

What does Llama 3.3 70B on RTX 5090 32GB cost per million tokens?

$0.1376 per 1M output tokens at 4× RTX 5090 32GB ($0.5000/hr). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Llama 3.3 70B need on RTX 5090 32GB?

70.6 GB for the weights at FP8 (E4M3), plus 21.5 GB of KV cache for 8 concurrent requests at 8K context — 92.1 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is RTX 5090 32GB the right choice for Llama 3.3 70B?

It works. Compare $0.1376 per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 8× RTX 4080 16GB at $0.1031, and the fastest is 4× Cerebras CS-3 (WSE-3) at 45076 tok/s.

Adjust the workload yourself

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

Open in calculator