Microsoft on Intel Battlemage

Can I run Phi-4 14B on an Arc B580?

Not at Q4_K_M — it needs 10.8 GB against 10.5 GB available. Drop to Q4_K_S and it fits, at about 26.4 tokens per second.

Won't fit

10.8 GB of 10.5 GB · 103%
011 GB
Weights 8.4 GB
KV cache 1.6 GB
Runtime overhead 0.8 GB
Over the limit 0.3 GB

Short by 0.3 GB. You can run it with 38 of 40 layers on the Arc B580 and the rest in system RAM, at roughly 19.2 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation19.2tok/s
Prompt processing657tok/s
Max context6Ktokens
KV per 1K tokens0GB

Every quantisation of Phi-4 14B on a Arc B580

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 27.4 GB 29.8 GB Won't fit 1.51
INT8 / W8A8 8.50 14.3 GB 16.7 GB Won't fit 4.07
Q8_0 (GGUF) 8.50 14.3 GB 16.7 GB Won't fit 4.07
FP8 (E4M3) 8.00 13.7 GB 16.1 GB Won't fit 4.43
Q6_K 6.56 11.2 GB 13.6 GB Won't fit 6.77
Q5_K_M 5.67 9.7 GB 12.1 GB Won't fit 10.6
Q5_K_S 5.52 9.4 GB 11.8 GB Won't fit 1K 11.7
AWQ 4-bit 4.25 8.7 GB 11.1 GB Won't fit 5K 16.7
GPTQ 4-bit 4.25 8.7 GB 11.1 GB Won't fit 5K 16.7
MXFP4 4.25 8.7 GB 11.1 GB Won't fit 5K 16.7
Q4_K_M 4.85 8.4 GB 10.8 GB Won't fit 6K 19.2
Q4_K_S 4.58 7.9 GB 10.3 GB Fits, but tight 9K 26.4
Q4_0 4.55 7.9 GB 10.3 GB Fits, but tight 9K 26.5
IQ4_XS 4.25 7.4 GB 9.8 GB Fits, but tight 11K 28.0
Q3_K_M 3.91 6.9 GB 9.3 GB Runs comfortably 14K 30.0
IQ3_M 3.70 6.5 GB 8.9 GB Runs comfortably 16K 31.4
IQ3_XXS 3.06 5.5 GB 7.9 GB Runs comfortably 16K 36.4
Q2_K 2.63 4.8 GB 7.2 GB Runs comfortably 16K 40.8
IQ2_XXS 2.06 3.9 GB 6.3 GB Runs comfortably 16K 48.5
IQ1_M 1.75 3.4 GB 5.8 GB Runs comfortably 16K 54.1

Also worth checking