TokenAssemble

Can an Intel Arc B580 run DeepSeek-R1?

No. DeepSeek-R1 needs 405.6GB at Q4_K_M with llama.cpp and 8K context; the Intel Arc B580 has 12GB of VRAM (11.5GB usable) — it does not fit, even with CPU offload.

Intel Arc B580 · DeepSeek-R1 · Q4_K_M

Won't fit

betaMLA memory formula — beta, not yet calibrated · New hardware class — not yet calibrated
weights 404.4 GBKV cache 0.58 GBoverhead 0.58 GBpool 11.5 GBneeds 405.6 GB — over the pool

assumes llama.cpp · batch 1 · 8K context · fp16 KV cache

No quant of this model fits cleanly on this hardware with llama-cpp at this context — a smaller model or more memory is the real fix.

$249 MSRP · as of 2026-07-08

Affiliate link — we may earn a commission. Verdicts are computed before any link is attached.

Every quant, graded

Intel Arc B580 × DeepSeek-R1, llama.cpp, 8K context.

QuantFile sizeFitSpeed tier
Q8_0713.3 GBWon't fit
Q6_K550.8 GBWon't fit
Q5_K_M475.4 GBWon't fit
Q4_K_M404.4 GBWon't fit

DeepSeek-R1 on similar hardware

More models on the Intel Arc B580

Verdict computed from published specs and measured GGUF file sizes — see how the math works.