Apple192 GB VRAM800 GB/s$0/hr

Qwen3.8 27B (VLM) on Mac M2 Ultra (192GB unified): Does It Fit, How Fast, What It Costs

Qwen3.8 27B (VLM) at FP8 (E4M3) needs 27.8 GB for its weights and 33.3 GB in total at 8K context with 8 concurrent users. That means it fits on a single Mac M2 Ultra (192GB unified). Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
33.3 GB needed / 192 GB available
Decode speed
28.08
tok/s per stream
Aggregate
243.0
tok/s @ 8 concurrent
Cost / 1M tokens
—
1× Mac M2 Ultra (192GB unified) · quote required
What changes as you add Mac M2 Ultra (192GB unified)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× Mac192 GB33.3 GB28.08243.033.78 s—✓
2× Mac384 GB33.3 GB47.73971.916.91 s—✓
4× Mac768 GB33.3 GB95.4638888.47 s—✓
8× Mac1536 GB33.3 GB190.9155504.25 s—✓
Concurrency vs context length
On 1× Mac M2 Ultra (192GB unified) 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?
4K3911.07 GB✓ fits
8K2382.15 GB✓ fits
33K718.59 GB✓ fits
66K3617.18 GB✓ fits
131K1834.36 GB✓ fits
Other GPUs that run Qwen3.8 27B (VLM)
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB2×2566$0.1083
AMD MI300X 192GB1×2232$0.3111
A100 40GB SXM41×652.8$0.3830
AMD MI325X 288GB1×2527$0.3847
A100 80GB SXM41×842.3$0.3957
Intel Gaudi 3 128GB1×1390$0.3997
Frequently asked questions

Can Qwen3.8 27B (VLM) run on Mac M2 Ultra (192GB unified)?

Yes. Qwen3.8 27B (VLM) at FP8 (E4M3) needs 33.3 GB including the KV cache, and Mac M2 Ultra (192GB unified) has 192 GB — so it fits on a single GPU.

How fast is Qwen3.8 27B (VLM) on Mac M2 Ultra (192GB unified)?

28.08 tokens/sec per stream and 243.0 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 33.78 s, and inter-token latency is 35.6 ms.

What does Qwen3.8 27B (VLM) on Mac M2 Ultra (192GB unified) cost per million tokens?

— per 1M output tokens at 1× Mac M2 Ultra (192GB unified). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does Qwen3.8 27B (VLM) need on Mac M2 Ultra (192GB unified)?

27.8 GB for the weights at FP8 (E4M3), plus 5.5 GB of KV cache for 8 concurrent requests at 8K context — 33.3 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is Mac M2 Ultra (192GB unified) the right choice for Qwen3.8 27B (VLM)?

It works. Compare — per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 4× RTX 3080 10GB at $0.0511, and the fastest is 1× Cerebras CS-3 (WSE-3) at 8423 tok/s.

Adjust the workload yourself

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

Open in calculator