Arc A580
Local inference speed
llama.cpp decode speed (tok/s) - estimated
| Workload | tok/s |
|---|---|
| LLaMA 3 8B Q4 | 59.9 |
Estimated - Estimated from memory bandwidth (llama.cpp decode) · updated 2026-11-05
Stacks for the Arc A580
Complete local setups that run on 8 GB
Prototype and deploy multi-agent RAG applications visually with local models
Unify multiple local and cloud models behind a single OpenAI-compatible API
Give your local AI superpowers by connecting it to files, web search, databases, and APIs via MCP
Arc A580Link to this section
The Arc A580 is an Intel GPU built on the Xe-HPG architecture, released 2023-10-10. For running AI locally, the numbers that matter are its 8 GB of GDDR6 and 512 GB/s of memory bandwidth. VRAM decides which models fit at all; bandwidth sets how fast they generate text.
What you can run on 8 GBLink to this section
At Q4_K_M quantization (the usual local default), 8 GB holds models up to roughly 12B parameters, leaving headroom for context. On this card you can run, among others:
- Darwin-9B-NEG - 9.7B parameters
- Qwen3.5 9B - 9.7B parameters
- NVIDIA-Nemotron-Nano-9B-v2 - 8.9B parameters
- LLaDA-8B-Instruct - 8B parameters
- LLaMA-Mesh - 8B parameters
Larger models need a higher-VRAM card, a second GPU, or CPU offload (which is much slower). Check any specific model with the VRAM calculator, or see the full picture on what can I run.
Local LLM speed (LLaMA 3, llama.cpp)Link to this section
Single-stream token-generation throughput - estimated from memory bandwidth:
| Model (quant) | Speed on Arc A580 |
|---|---|
| Llama 3 8B (Q4_K_M) | 59.9 tok/s |
| Llama 3 8B (F16) | ✗ won't fit |
| Llama 3 70B (Q4_K_M) | ✗ won't fit |
Because decode is memory-bandwidth bound, the 512 GB/s figure is the best single predictor of chat speed on this card. Estimates are calibrated against measured RTX-40-series cards and are typically within ~15%.
Memory and powerLink to this section
- VRAM: 8 GB GDDR6 (256-bit bus)
- Bandwidth: 512 GB/s
- TDP: 175 W - a 450 W+ power supply is recommended
- Process: 6 nm
- Interface: PCIe 4.0 x16
Quantization and contextLink to this section
Quantization trades a little quality for a lot of VRAM. On 8 GB you can fit roughly a 12B model at Q4_K_M, about a 6B model at the higher-quality Q8, or a smaller model at full FP16. Longer context windows also consume VRAM (the KV cache grows with context length), so leave a few GB of headroom if you plan to use large prompts or many concurrent requests. For most chat and coding use, Q4_K_M on this card is the sweet spot between speed, quality, and the 8 GB budget.
How it comparesLink to this section
Similar cards for local AI, by VRAM and 8B-Q4 speed:
| GPU | VRAM | Bandwidth | Llama 3 8B Q4 |
|---|---|---|---|
| Arc A580 | 8 GB | 512 GB/s | 59.9 tok/s |
| AMD Radeon RX 7600 | 8 GB | 288 GB/s | 33.7 tok/s |
| Intel Arc A750 | 8 GB | 512 GB/s | 59.9 tok/s |
| NVIDIA RTX 4060 Ti 8GB | 8 GB | 288 GB/s | 33.7 tok/s |
Bottom lineLink to this section
The Arc A580 is best for entry-level, budget-llm. With under 12 GB, stick to small quantized models (up to ~7B). If you need more, compare with AMD Radeon RX 7600 and Intel Arc A750.
SourcesLink to this section
- Specifications: RightNow GPU Database (TechPowerUp data)
- Benchmarks: GPU-Benchmarks-on-LLM-Inference (basis for the bandwidth estimate)
Specs and benchmarks last checked 2026-06-08. Verify current pricing before buying.
Frequently asked
Quick answers to common questions
How much VRAM does the Arc A580 have?
The Arc A580 has 8 GB of VRAM with 512 GB/s memory bandwidth.
What local AI models can run on the Arc A580?
The Arc A580 with 8 GB VRAM can run many models depending on quantization. Models up to ~12B params may fit at Q4_K_M. Use our VRAM calculator to check specific models.
Is the Arc A580 good for local AI inference?
Arc A580 is best for entry-level, budget-llm. Check our hardware directory for alternatives with more VRAM.
Where can I buy the Arc A580?
Check our buy links above for the best current prices on Amazon, Newegg, and B&H. Prices vary by retailer and availability.
How does the Arc A580 compare to other GPUs?
Arc A580 has 8 GB VRAM and 512 GB/s bandwidth. It works best with smaller quantized models. Browse our hardware directory for side-by-side comparisons.
What power supply do I need for the Arc A580?
The Arc A580 has a TDP of 175W. A standard quality PSU of 650W+ should suffice. Always check the manufacturer's recommendations for your specific build.
Nearby options
Similar hardware and models that fit