Can Aya Expanse 32B run on AMD Instinct MI250X 128GB?

YES — Runs Great

C53Usable

Estimated from fit model

Aya Expanse 32B needs ~35.7 GB VRAM. AMD Instinct MI250X 128GB has 128.0 GB. With Q4_K_M quantization, expect ~139 tok/s.

Runtime: llama.cppCapacity: RoomyBandwidth: HighStack: StandardBottleneck: Balanced

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 35.7 GB, 139.1 tok/s, Runs well

35.7 GB required128.0 GB available

28% VRAM used

Fit status

Runs well

Decode

139.1 tok/s

TTFT

1392 ms

Safe context

Memory

35.7 GB / 128.0 GB

Memory breakdown

Weights19.5 GB

KV Cache2.4 GB

Runtime0.9 GB

Headroom12.8 GB

See how fast it feels

See how fast it feelsAya Expanse 32B on AMD Instinct MI250X 128GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API.

When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed.

Key benefits of running locally:
  - Full privacy: your prompts never leave your machine
  - No per-token cost or rate limits
  - Works offline once the model is downloaded
  - Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb:

  Model size     Q4 (4-bit)    Q8 (8-bit)    FP16
  7B params      ~4.3 GB       ~7.5 GB       ~14 GB
  13B params     ~7.9 GB       ~13.9 GB      ~26 GB
  70B params     ~42.7 GB      ~74.9 GB      ~140 GB

Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 139.1 tok/s decode · 1.4s TTFT (warm) · 348 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

No major red flags

This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	C	Runs well	139.1 tok/s	759 ms	8K
Coding	C	Runs well	139.1 tok/s	1392 ms	8K
Agentic Coding	C	Runs well	139.1 tok/s	2025 ms	8K
Reasoning	C	Runs well	139.1 tok/s	1645 ms	8K
RAG	C	Runs well	139.1 tok/s	2531 ms	8K

Inference speed

Aya Expanse 32B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Aya Expanse 32B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~43 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	43.3	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	33.5	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	33.5	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	31.0	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	25.9	Offloads
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	25.9	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	24.5	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	21.1	Fits
RTX 4090 24GB	24 GB	Q4_K_M	16.6	Offloads
RTX 3090 24GB	24 GB	Q4_K_M	15.2	Offloads
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	13.4	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	12.3	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	10.2	Too big
RTX 4070 12GB	12 GB	Q4_K_M	3.6	Too big
RTX 3060 12GB	12 GB	Q4_K_M	2.2	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.0	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How Aya Expanse 32B (32B params) fits at each quantization level on AMD Instinct MI250X 128GB (128.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	12.5 GB	Low	C44
Q3_K_S	3	15.7 GB	Low	C44
NVFP4	4	17.9 GB	Medium	C44
Q4_K_M	4	19.5 GB	Medium	C44
Q5_K_M	5	23.0 GB	High	C45
Q6_K	6	26.2 GB	High	C45
Q8_0	8	34.2 GB	Very High	C47
F16Best for your GPU	16	65.6 GB	Maximum	C52

Get started

Copy-paste commands to run Aya Expanse 32B on your machine.

Run

ollama run aya-expanse:32b

Frequently asked questions

Can AMD Instinct MI250X 128GB run Aya Expanse 32B?

Yes, AMD Instinct MI250X 128GB can run Aya Expanse 32B with a C grade (Runs well). Expected decode speed: 139.1 tok/s.

How much VRAM does Aya Expanse 32B need?

Aya Expanse 32B (32B parameters) requires approximately 35.7 GB of memory with Q4_K_M quantization.

What is the best quantization for Aya Expanse 32B?

The recommended quantization for Aya Expanse 32B is Q4_K_M, which balances quality and memory efficiency.

What speed will Aya Expanse 32B run at on AMD Instinct MI250X 128GB?

On AMD Instinct MI250X 128GB, Aya Expanse 32B achieves approximately 139.1 tokens per second decode speed with a time-to-first-token of 1392ms using Q4_K_M quantization.

Can AMD Instinct MI250X 128GB run Aya Expanse 32B for coding?

For coding workloads, Aya Expanse 32B on AMD Instinct MI250X 128GB receives a C grade with 139.1 tok/s and 8K context.

What context window can Aya Expanse 32B use on AMD Instinct MI250X 128GB?

On AMD Instinct MI250X 128GB, Aya Expanse 32B can safely use up to 8K tokens of context. The model's official context limit is 8K, but available memory constrains the safe maximum.

See all results for AMD Instinct MI250X 128GB See all hardware for Aya Expanse 32B

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/aya-expanse-32b-on-instinct-mi250x-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: