Can Apertus v1.5 8B run on Intel Arc A580 8GB?

BARELY — Tight on Memory

B65Good

Estimated from fit model

Apertus v1.5 8B needs ~9.1 GB VRAM. Intel Arc A580 8GB has 8.0 GB. With Q4_K_M quantization, expect ~29 tok/s.

Runtime: llama.cppCapacity: OffloadBandwidth: MediumStack: StandardBottleneck: Host offload

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) — 9.1 GB, 28.5 tok/s, Very compromised (needs ~0.6 GB host RAM)

9.1 GB required8.0 GB available

114% VRAM needed

1.1 GB over capacity — needs offload or smaller quantization

Fit status

Very compromised (needs ~0.6 GB host RAM)

Decode

28.5 tok/s

TTFT

6787 ms

Safe context

Memory

9.1 GB / 8.0 GB

Offload

10%

Memory breakdown

Weights5.4 GB

KV Cache2.0 GB

Runtime0.9 GB

Headroom0.8 GB

See how fast it feels

See how fast it feelsApertus v1.5 8B on Intel Arc A580 8GB
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: 28.5 tok/s decode · 6.8s TTFT (warm) · 71 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Runtime ecosystem is narrower than CUDA

Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.

Best improvement path

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Prefer CUDA if you want the path of least resistance

If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs with offload (needs ~0.1 GB host RAM)	36.2 tok/s	2914 ms	7K
Coding	B	Very compromised (needs ~0.6 GB host RAM)	28.5 tok/s	6787 ms	7K
Agentic Coding	F	Too heavy	18.9 tok/s	14875 ms	7K
Reasoning	B	Very compromised (needs ~0.6 GB host RAM)	28.5 tok/s	8021 ms	7K
RAG	F	Too heavy	18.9 tok/s	18594 ms	7K

Inference speed

Apertus v1.5 8B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Apertus v1.5 8B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~169 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	169.1	Fits
RTX 4090 24GB	24 GB	Q4_K_M	139.6	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	124.6	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	111.3	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	110.3	Fits
RTX 3090 24GB	24 GB	Q4_K_M	106.8	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	91.9	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	87.1	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	74.2	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	74.2	Fits
RTX 4070 12GB	12 GB	Q4_K_M	68.9	Fits
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	47.5	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	43.6	Fits
RTX 3060 12GB	12 GB	Q4_K_M	39.5	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	38.3	Fits
RTX 4060 8GB	8 GB	Q4_K_M	20.8	Heavy offload

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 Apertus v1.5 8B (8.899999618530273B params) fits at each quantization level on Intel Arc A580 8GB (8.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q1_0_G128	1.125	1.3 GB	Very Low	A75
Q2_0_G128	1.71	2.4 GB	Low	A77
Q2_K	2	3.5 GB	Low	A78
Q3_K_S	3	4.4 GB	Low	A78
NVFP4	4	5.0 GB	Medium	A77
Q4_K_MBest for your GPU	4	5.4 GB	Medium	A77
Q5_K_M	5	6.4 GB	High	F0
Q6_K	6	7.3 GB	High	F0
Q8_0	8	9.5 GB	Very High	F0
F16	16	18.2 GB	Maximum	F0

Get started

Copy-paste commands to run Apertus v1.5 8B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "swiss-ai/Apertus-v1.5-8B" \
  --hf-file "Apertus-v1.5-8B-Q4_K_M.gguf" \
  -c 4096 -ngl 99

Opciones de mejora

Hardware que ejecuta bien Apertus v1.5 8B

Intel Arc B570 10GBOpción económica

10 GB VRAM (+2)

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.40.6 tok/s decodificación

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

Sube la velocidad estimada de decodificación alrededor de un 42%.

~$219 MSRP

Intel Arc B580 12GBMejor relación calidad-precio

12 GB VRAM (+4)

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.43.3 tok/s decodificación

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

Sube la velocidad estimada de decodificación alrededor de un 52%.

~$249 MSRP

Intel Arc A770 16GBMejora Intel

16 GB VRAM (+8)560 GB/s (+48)

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.49.9 tok/s decodificación

Elimina el offload a memoria del sistema, que suele ser la mayor mejora individual en latencia y throughput.

Sube la velocidad estimada de decodificación alrededor de un 75%.

~$349 MSRP

Frequently asked questions

Can Intel Arc A580 8GB run Apertus v1.5 8B?

Yes, Intel Arc A580 8GB can run Apertus v1.5 8B with a B grade (Very compromised (needs ~0.6 GB host RAM)). Expected decode speed: 28.5 tok/s.

How much VRAM does Apertus v1.5 8B need?

Apertus v1.5 8B (8.899999618530273B parameters) requires approximately 9.1 GB of memory with Q4_K_M quantization.

What is the best quantization for Apertus v1.5 8B?

The recommended quantization for Apertus v1.5 8B is Q4_K_M, which balances quality and memory efficiency.

What speed will Apertus v1.5 8B run at on Intel Arc A580 8GB?

On Intel Arc A580 8GB, Apertus v1.5 8B achieves approximately 28.5 tokens per second decode speed with a time-to-first-token of 6787ms using Q4_K_M quantization.

Can Intel Arc A580 8GB run Apertus v1.5 8B for coding?

For coding workloads, Apertus v1.5 8B on Intel Arc A580 8GB receives a B grade with 28.5 tok/s and 7K context.

What context window can Apertus v1.5 8B use on Intel Arc A580 8GB?

On Intel Arc A580 8GB, Apertus v1.5 8B can safely use up to 7K tokens of context. The model's official context limit is 262K, but available memory constrains the safe maximum.

What should I upgrade first if Apertus v1.5 8B feels slow on Intel Arc A580 8GB?

Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Would CUDA be a better path than Intel Arc A580 8GB for Apertus v1.5 8B?

Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.

See all results for Intel Arc A580 8GB See all hardware for Apertus v1.5 8B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/apertus-v1.5-8b-on-arc-a580-8gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: