Google95 GB VRAM819 GB/s$4.2/hr

Qwen2 72B on TPU v5p (per-chip, 95GB): Does It Fit, How Fast, What It Costs

Qwen2 72B at FP8 (E4M3) needs 72.7 GB for its weights and 94.2 GB in total at 8K context with 8 concurrent users. That means it fits on a single TPU v5p (per-chip, 95GB). Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
94.2 GB needed / 95 GB available
Decode speed
10.98
tok/s per stream
Aggregate
131.8
tok/s @ 8 concurrent
Cost / 1M tokens
$8.85
1× TPU v5p (per-chip, 95GB)
What changes as you add TPU v5p (per-chip, 95GB)s
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× TPU95 GB94.2 GB10.98131.85.31 s$8.85✓
2× TPU190 GB94.2 GB18.67448.12.68 s$5.21✓
4× TPU380 GB94.2 GB37.3517931.35 s$2.60✓
8× TPU760 GB94.2 GB74.697170692.1 ms$1.30✓
Concurrency vs context length
On 1× TPU v5p (per-chip, 95GB) 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?
4K161.34 GB✓ fits
8K82.68 GB✓ fits
33K210.74 GB✓ fits
66K121.47 GB✓ fits
131K042.95 GB✗ does not fit
Other GPUs that run Qwen2 72B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB4×3922$0.1416
A100 80GB SXM42×1094$0.6092
H100 SXM5 80GB2×1833$0.7577
AMD MI300X 192GB1×853.0$0.8142
L40S 48GB2×472.8$0.8226
AMD MI325X 288GB1×965.6$1.01
Frequently asked questions

Can Qwen2 72B run on TPU v5p (per-chip, 95GB)?

Yes. Qwen2 72B at FP8 (E4M3) needs 94.2 GB including the KV cache, and TPU v5p (per-chip, 95GB) has 95 GB — so it fits on a single GPU.

How fast is Qwen2 72B on TPU v5p (per-chip, 95GB)?

10.98 tokens/sec per stream and 131.8 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 5.31 s, and inter-token latency is 91.0 ms.

What does Qwen2 72B on TPU v5p (per-chip, 95GB) cost per million tokens?

$8.85 per 1M output tokens at 1× TPU v5p (per-chip, 95GB) ($4.20/hr). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Qwen2 72B need on TPU v5p (per-chip, 95GB)?

72.7 GB for the weights at FP8 (E4M3), plus 21.5 GB of KV cache for 8 concurrent requests at 8K context — 94.2 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is TPU v5p (per-chip, 95GB) the right choice for Qwen2 72B?

It works. Compare $8.85 per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 8× RTX 4080 16GB at $0.1062, and the fastest is 4× Cerebras CS-3 (WSE-3) at 43774 tok/s.

Adjust the workload yourself

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

Open in calculator