Qwen on NVIDIA Ampere

Can I run Qwen2.5 72B on an RTX 3090?

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 RTX 3090 and the rest in system RAM, at roughly 1.59 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation1.59tok/s
Prompt processing205tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen2.5 72B on a RTX 3090

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 — 0.33
INT8 / W8A8 8.50 71.4 GB 74.8 GB Won't fit — 0.72
Q8_0 (GGUF) 8.50 71.4 GB 74.8 GB Won't fit — 0.72
FP8 (E4M3) 8.00 67.7 GB 71.1 GB Won't fit — 0.77
Q6_K 6.56 55.5 GB 58.9 GB Won't fit — 1.01
Q5_K_M 5.67 48.0 GB 51.4 GB Won't fit — 1.25
Q5_K_S 5.52 46.7 GB 50.1 GB Won't fit — 1.30
Q4_K_M 4.85 41.2 GB 44.6 GB Won't fit — 1.59
AWQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit — 1.73
GPTQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit — 1.73
MXFP4 4.25 39.4 GB 42.8 GB Won't fit — 1.73
Q4_K_S 4.58 39.0 GB 42.4 GB Won't fit — 1.74
Q4_0 4.55 38.8 GB 42.2 GB Won't fit — 1.75
IQ4_XS 4.25 36.3 GB 39.7 GB Won't fit — 1.99
Q3_K_M 3.91 33.6 GB 36.9 GB Won't fit — 2.30
IQ3_M 3.70 31.8 GB 35.2 GB Won't fit — 2.60
IQ3_XXS 3.06 26.6 GB 30.0 GB Won't fit — 3.98
Q2_K 2.63 23.1 GB 26.5 GB Won't fit — 6.15
IQ2_XXS 2.06 18.4 GB 21.8 GB Won't fit 8K 25.1
IQ1_M 1.75 15.9 GB 19.3 GB Runs comfortably 16K 33.6

Also worth checking

Qwen2.5 72B on every GPU
Full hardware tableEverything that runs on a RTX 3090
Full model tableWhy the KV cache is the part that bites
Guide