Cerebras44 GB VRAM20,000 GB/s

Qwen 3 4B on Cerebras CS-3 (WSE-3): Does It Fit, How Fast, What It Costs

Qwen 3 4B at FP16 needs 8.0 GB for its weights and 17.7 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
17.7 GB needed / 44 GB available
Decode speed
1625
tok/s per stream
Aggregate
19500
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 GB17.7 GB162519500135.5 ms—✓
2× Cerebras88 GB17.7 GB27636630082.8 ms—✓
4× Cerebras176 GB17.7 GB552526520056.4 ms—✓
8× Cerebras352 GB17.7 GB11050106080043.2 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?
4K59618.5 MB✓ fits
8K291.21 GB✓ fits
33K74.83 GB✓ fits
66K39.66 GB✓ fits
131K119.33 GB✓ fits
Other GPUs that run Qwen 3 4B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 3090 24GB1×912.6$0.0761
RTX 5090 32GB1×1747$0.0795
RTX 4090 24GB1×982.8$0.1131
AMD MI300X 192GB1×5168$0.1344
V100 SXM2 32GB1×877.5$0.1583
A100 40GB SXM41×1511$0.1654
Frequently asked questions

Can Qwen 3 4B run on Cerebras CS-3 (WSE-3)?

Yes. Qwen 3 4B at FP16 needs 17.7 GB including the KV cache, and Cerebras CS-3 (WSE-3) has 44 GB — so it fits on a single GPU.

How fast is Qwen 3 4B on Cerebras CS-3 (WSE-3)?

1625 tokens/sec per stream and 19500 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 135.5 ms, and inter-token latency is 615.4 µs.

What does Qwen 3 4B 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 Qwen 3 4B need on Cerebras CS-3 (WSE-3)?

8.0 GB for the weights at FP16, plus 9.7 GB of KV cache for 8 concurrent requests at 8K context — 17.7 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is Cerebras CS-3 (WSE-3) the right choice for Qwen 3 4B?

It works. Compare — per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 2× RTX 3080 10GB at $0.0441, and the fastest is 1× Cerebras CS-3 (WSE-3) at 19500 tok/s.

Adjust the workload yourself

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

Open in calculator