AMD288 GB VRAM6,000 GB/s$3.5/hr

GPT-NeoX 20B on AMD MI325X 288GB: Does It Fit, How Fast, What It Costs

GPT-NeoX 20B at FP8 (E4M3) needs 20.0 GB for its weights and 90.9 GB in total at 8K context with 8 concurrent users. That means it fits on a single AMD MI325X 288GB. Every figure is computed by tokcalc's open-source capacity model.

Fits?
Yes, 1 GPU
90.9 GB needed / 288 GB available
Decode speed
292.5
tok/s per stream
Aggregate
3510
tok/s @ 8 concurrent
Cost / 1M tokens
$0.2770
1× AMD MI325X 288GB
What changes as you add AMD MI325X 288GBs
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× AMD288 GB90.9 GB292.53510534.8 ms$0.2770✓
2× AMD576 GB90.9 GB497.311934282.7 ms$0.1629✓
4× AMD1152 GB90.9 GB994.547736156.4 ms$0.0815✓
8× AMD2304 GB90.9 GB198919094493.2 ms$0.0407✓
Concurrency vs context length
On 1× AMD MI325X 288GB 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?
4K604.43 GB✓ fits
8K308.86 GB✓ fits
33K735.43 GB✓ fits
66K370.87 GB✓ fits
131K1141.73 GB✓ fits
Other GPUs that run GPT-NeoX 20B
Same workload, same quantization — ranked by value (tokens per dollar-hour).
GPUGPUsAggregate tok/sCost / 1M tok
RTX 5090 32GB4×14257$0.0390
A100 80GB SXM42×3978$0.1676
H100 SXM5 80GB2×6663$0.2084
AMD MI300X 192GB1×3101$0.2240
L40S 48GB2×1718$0.2263
Intel Gaudi 3 128GB1×1931$0.2878
Frequently asked questions

Can GPT-NeoX 20B run on AMD MI325X 288GB?

Yes. GPT-NeoX 20B at FP8 (E4M3) needs 90.9 GB including the KV cache, and AMD MI325X 288GB has 288 GB — so it fits on a single GPU.

How fast is GPT-NeoX 20B on AMD MI325X 288GB?

292.5 tokens/sec per stream and 3510 tok/s aggregate with continuous batching at 8 concurrent users. Time to first token is 534.8 ms, and inter-token latency is 3.4 ms.

What does GPT-NeoX 20B on AMD MI325X 288GB cost per million tokens?

$0.2770 per 1M output tokens at 1× AMD MI325X 288GB ($3.50/hr). That is the marginal hardware cost — add idle capacity, ops, and egress for the real figure.

How much VRAM does GPT-NeoX 20B need on AMD MI325X 288GB?

20.0 GB for the weights at FP8 (E4M3), plus 70.9 GB of KV cache for 8 concurrent requests at 8K context — 90.9 GB total. Every extra concurrent user at long context costs more, so leave headroom.

Is AMD MI325X 288GB the right choice for GPT-NeoX 20B?

It works. Compare $0.2770 per 1M tokens against the alternatives below — the cheapest option that fits the same workload is 8× RTX 4080 16GB at $0.0292, and the fastest is 4× Cerebras CS-3 (WSE-3) at 159120 tok/s.

Adjust the workload yourself

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

Open in calculator