Qwen on AMD RDNA3

Can I run Qwen2.5 72B on an Radeon RX 7900 XTX?

Not at Q4_K_M — it needs 44.6 GB against 21.8 GB available. You would need 3 of these cards.

Won't fit

44.6 GB of 21.8 GB · 205%
046 GB
Weights 41.2 GB
KV cache 2.5 GB
Runtime overhead 0.9 GB
Over the limit 22.9 GB

Short by 22.9 GB. You can run it with 35 of 80 layers on the Radeon RX 7900 XTX and the rest in system RAM, at roughly 1.44 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation1.44tok/s
Prompt processing355tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen2.5 72B on a Radeon RX 7900 XTX

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 135.4 GB 138.8 GB Won't fit 0.30
INT8 / W8A8 8.50 71.4 GB 74.8 GB Won't fit 0.65
Q8_0 (GGUF) 8.50 71.4 GB 74.8 GB Won't fit 0.65
FP8 (E4M3) 8.00 67.7 GB 71.1 GB Won't fit 0.70
Q6_K 6.56 55.5 GB 58.9 GB Won't fit 0.92
Q5_K_M 5.67 48.0 GB 51.4 GB Won't fit 1.13
Q5_K_S 5.52 46.7 GB 50.1 GB Won't fit 1.19
Q4_K_M 4.85 41.2 GB 44.6 GB Won't fit 1.44
AWQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit 1.57
GPTQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit 1.57
MXFP4 4.25 39.4 GB 42.8 GB Won't fit 1.57
Q4_K_S 4.58 39.0 GB 42.4 GB Won't fit 1.59
Q4_0 4.55 38.8 GB 42.2 GB Won't fit 1.60
IQ4_XS 4.25 36.3 GB 39.7 GB Won't fit 1.81
Q3_K_M 3.91 33.6 GB 36.9 GB Won't fit 2.09
IQ3_M 3.70 31.8 GB 35.2 GB Won't fit 2.37
IQ3_XXS 3.06 26.6 GB 30.0 GB Won't fit 3.63
Q2_K 2.63 23.1 GB 26.5 GB Won't fit 5.62
IQ2_XXS 2.06 18.4 GB 21.8 GB Won't fit 8K 23.3
IQ1_M 1.75 15.9 GB 19.3 GB Runs comfortably 16K 31.4

Also worth checking