Skip to content
Run LLM Locally

Qwen3 4B on the Arc B580

With 12 GB of VRAM, the Arc B580 runs Qwen3 4B at Q4_K_M entirely on the GPU with an 8K context (about 4.6 GB needed). At Q8_0 it fits entirely.

ShareFacebookXWhatsAppLinkedInEmail

Qwen3 4B

Q4_K_M · 8K context

Excellent

The model needs about 4.6 GB and fits in the 11.6 GB available, with 7.1 GB to spare.

Your hardware
Arc B580 12 GB · 64 GB RAM
Estimated memory
~4.6 GB estimate
VRAM (usable memory)
11.6 GB
GPU only
Yes
CPU/GPU offload
Not required
Generation speed
55–87 tok/s estimate
Largest context in fast memory
56K tokens
Confidence: low

Based on: memory bandwidth (456 GB/s) and model size

Memory breakdown

~4.6 GB (4.2 GB–5 GB) needed · estimate

Fits entirely in VRAM · 7.1 GB left

  • Model weights2.3 GB
  • KV cache1.1 GB
  • Runtime overhead0.8 GB
  • Safety margin0.3 GB
Assumptions: context 8K tokens · KV cache f16 · llama.cpp (SYCL) · flash attention on · weights = real file size

EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card

What makes this estimate less reliable

  • no benchmark for this hardware and model yet

Other quantizations on the Arc B580

Qwen3 4B on Arc B580 + 64 GB RAM · 8K context
Qwen3 4B Q8_08K~6.2 GB · 35–55 tok/s estimateExcellent
Qwen3 4B Q6_K8K~5.3 GB · 44–69 tok/s estimateExcellent
Qwen3 4B Q5_K_M8K~4.9 GB · 49–77 tok/s estimateExcellent
Qwen3 4B Q4_K_M8K~4.6 GB · 55–87 tok/s estimateExcellent
Qwen3 4B IQ4_XS8K~4.4 GB · 60–94 tok/s estimateExcellent
Qwen3 4B Q3_K_M8K~4.2 GB · 64–100 tok/s estimateExcellent

Alternatives

GPUs that run Qwen3 4B Q4_K_M entirely in VRAM

Other models that fit the Arc B580

Where to buy a Arc B580

Prices and links follow the market you choose.

Intel Arc B580

12 GB GDDR6 · 456 GB/s · 190 W

ExcellentRuns Qwen3 4B Q4_K_M entirely in memory

Price not available

Availability unknown

View on amazon.com

Affiliate link

No current offer imported for this market: the link opens a marketplace search.

Frequently asked questions

Can the Arc B580 run Qwen3 4B?
The model needs about 4.6 GB and fits in the 11.6 GB available, with 7.1 GB to spare.
Which quantization should I use on the Arc B580?
Q8_0 is the highest-precision quantization that fits entirely in VRAM at 8K context. ~6.2 GB / 11.6 GB.
What is the largest context on the Arc B580?
At Q4_K_M, about 56K tokens fit entirely in VRAM with an f16 KV cache. A q8_0 KV cache roughly doubles that.
How fast is Qwen3 4B (Arc B580)?
We estimate 55–87 tok/s (estimate, confidence low). Based on: memory bandwidth (456 GB/s) and model size.