Cerebras44 GB VRAM20,000 GB/s

Gemma 2 9B on Cerebras CS-3 (WSE-3): Does It Fit, How Fast, What It Costs

Gemma 2 9B at FP16 needs 18.0 GB for its weights and 40.5 GB in total at 8K context with 8 concurrent users. That means it fits on a single Cerebras CS-3 (WSE-3). Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
40.5 GB needed / 44 GB available
Decode speed
722.2
tok/s per stream
Aggregate
8667
tok/s @ 8 concurrent
Cost / 1M tokens
—
1× Cerebras CS-3 (WSE-3) · quote required
What changes as you add Cerebras CS-3 (WSE-3)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× Cerebras44 GB40.5 GB722.28667267.3 ms—✓
2× Cerebras88 GB40.5 GB122829467148.8 ms—✓
4× Cerebras176 GB40.5 GB245611786789.4 ms—✓
8× Cerebras352 GB40.5 GB491147146759.7 ms—✓
Concurrency vs context length
On 1× Cerebras CS-3 (WSE-3) at FP16. KV cache scales linearly with context, so this is the number of users your GPUs can actually serve.
ContextConcurrent usersKV per requestFits?
4K181.41 GB✓ fits
8K92.82 GB✓ fits
33K211.27 GB✓ fits
66K122.55 GB✓ fits
131K045.10 GB✗ does not fit
Other GPUs that run Gemma 2 9B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB2×2640$0.1052
AMD MI300X 192GB1×2297$0.3024
AMD MI325X 288GB1×2600$0.3739
A100 80GB SXM41×866.7$0.3846
Intel Gaudi 3 128GB1×1430$0.3885
H100 SXM5 80GB1×1452$0.4784
Frequently asked questions

Can Gemma 2 9B run on Cerebras CS-3 (WSE-3)?

Yes. Gemma 2 9B at FP16 needs 40.5 GB including the KV cache, and Cerebras CS-3 (WSE-3) has 44 GB — so it fits on a single GPU.

How fast is Gemma 2 9B on Cerebras CS-3 (WSE-3)?

722.2 tokens/sec per stream and 8667 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 267.3 ms, and inter-token latency is 1.4 ms.

What does Gemma 2 9B on Cerebras CS-3 (WSE-3) cost per million tokens?

— per 1M output tokens at 1× Cerebras CS-3 (WSE-3). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Gemma 2 9B need on Cerebras CS-3 (WSE-3)?

18.0 GB for the weights at FP16, plus 22.5 GB of KV cache for 8 concurrent requests at 8K context — 40.5 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is Cerebras CS-3 (WSE-3) the right choice for Gemma 2 9B?

It works. Compare — per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 8× RTX 3080 10GB at $0.0248, and the fastest is 8× RTX 3080 10GB at 17916 tok/s.

Adjust the workload yourself

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

Open in calculator