How much VRAM does the Intel Arc A580 8GB have?

The Intel Arc A580 8GB has 8 GB of GDDR6 with 512 GB/s memory bandwidth.

What is the Intel Arc A580 8GB best for?

With 8 GB of VRAM, the Intel Arc A580 8GB is best for running compact models (1B–8B) at low quantization, suitable for edge inference, prototyping, and lightweight tasks.

What LLMs can the Intel Arc A580 8GB run locally?

The Intel Arc A580 8GB can run 26 of the 94 open-weight models tracked by CanItRun natively in VRAM at 8k context. Top options include: Ornith 1.5 9B at Q5_K_M, Qwen 3.5 9B at Q5_K_M, Bonsai 27B at 1-bit (Q1_0).

Can the Intel Arc A580 8GB run Gemma 4 31B?

The Intel Arc A580 8GB can run Gemma 4 31B with CPU offload at Q6_K quantization, but inference will be slower than native VRAM execution.

Can the Intel Arc A580 8GB run Qwen 3.6 27B?

The Intel Arc A580 8GB can run Qwen 3.6 27B with CPU offload at Q8_0 quantization, but inference will be slower than native VRAM execution.

Can the Intel Arc A580 8GB run Qwen3 8B?

Yes. The Intel Arc A580 8GB runs Qwen3 8B natively in VRAM at Q4_K_M quantization, achieving approximately 54.7 tokens per second.

Intel Arc A580 8GB

The Intel Arc A580 8GB has 8 GB VRAM and 512 GB/s memory bandwidth. It can run 26 of our 94 tracked models natively in VRAM at 8k context.

With 8 GB GDDR6, the Intel Arc A580 8GB is a consumer-tier GPU that can run 26 models natively. It's best for smaller models under 8B parameters.

Intel Arc A580 8GB: 2023 Xe-HPG Alchemist with 8GB GDDR6 at 384 GB/s, budget Alchemist with 8GB VRAM.

7B at Q4 natively; lower bandwidth than A750/A770 reduces token rates. ~3-5 t/s for 7B via Vulkan.

Vulkan via llama.cpp works cross-platform. SYCL backend requires oneAPI toolkit. Ollama support limited.

Vendor	Intel
Architecture	Xe-HPG (Alchemist)
VRAM	8 GB
Memory type	GDDR6
Memory bandwidth	512 GB/s
Compute backend	VULKAN
Tier	Consumer
Released	2023
Models (native)	26 / 94
Models (offload)	30 / 94

Software: Vulkan backend works in llama.cpp; SYCL backend available but requires oneAPI toolkit. Ollama support is limited.

Popular models for this GPU

Bonsai 27B Phi-4 14B Instruct Mistral Nemo 12B Instruct Gemma 3 12B Instruct Gemma 2 9B Instruct

Models this GPU runs natively in VRAM (26)

Show 21 more

Models that fit with CPU offload (30)

These use system RAM for layers that don't fit in VRAM, so expect much slower inference.

Too large for this GPU (38)

Explore

Browse all guides →Browse all models →Browse all GPUs →

Frequently asked questions

How much VRAM does the Intel Arc A580 8GB have?: The Intel Arc A580 8GB has 8 GB of GDDR6 with 512 GB/s memory bandwidth.
What is the Intel Arc A580 8GB best for?: With 8 GB of VRAM, the Intel Arc A580 8GB is best for running compact models (1B–8B) at low quantization, suitable for edge inference, prototyping, and lightweight tasks.
What LLMs can the Intel Arc A580 8GB run locally?: The Intel Arc A580 8GB can run 26 of the 94 open-weight models tracked by CanItRun natively in VRAM at 8k context. Top options include: Ornith 1.5 9B at Q5_K_M, Qwen 3.5 9B at Q5_K_M, Bonsai 27B at 1-bit (Q1_0).
Can the Intel Arc A580 8GB run Gemma 4 31B?: The Intel Arc A580 8GB can run Gemma 4 31B with CPU offload at Q6_K quantization, but inference will be slower than native VRAM execution.
Can the Intel Arc A580 8GB run Qwen 3.6 27B?: The Intel Arc A580 8GB can run Qwen 3.6 27B with CPU offload at Q8_0 quantization, but inference will be slower than native VRAM execution.
Can the Intel Arc A580 8GB run Qwen3 8B?: Yes. The Intel Arc A580 8GB runs Qwen3 8B natively in VRAM at Q4_K_M quantization, achieving approximately 54.7 tokens per second.